Early tracking of lateral objects

Abstract

A method for tracking a neighboring vehicle in the environment of a main vehicle, implemented by a driving assistance device of the main vehicle and connected to a camera capable of acquiring images within a field of view and at acquisition times, the method comprising: - detecting, from a first image acquired at a first acquisition time, a first object lateral to the main vehicle, - determining data associated with the first detected object, - creating a virtual object associated with parameters including a current position associated with the virtual object, said current position being estimated from said data, - estimating, from said current position, a future position of the virtual object, - updating the parameters from said future position with, if a first criterion is met, an update of the current length of the virtual object. Abstract figure: Figure 9

Description

Title of the invention: Early tracking of lateral objects technical field

[0001] This disclosure falls within the field of driver assistance systems and more particularly the automatic detection of vehicles in areas of interest, including lateral areas. Previous technique

[0002] The rise of Intelligent Transport Systems (ITS) has led to the development of numerous embedded systems in vehicles, particularly road transport vehicles. Such embedded systems include, in particular, driver assistance systems and autonomous driving systems. Specifically, the detection and / or tracking of objects, including vehicles, plays an important role in issues such as traffic flow, road safety, and the management of road infrastructure (for example, variable message signs or automated speed cameras).

[0003] In the context of embedded systems in autonomous or semi-autonomous vehicles operating in road traffic (for example, on a highway), the early detection and tracking of objects surrounding the vehicle are particularly important in order to avoid collisions between the vehicle and other vehicles. In particular, in lane change assistance or automatic lane change systems, early detection and reliable tracking of obstacles such as a median strip or another vehicle in a lane adjacent to the vehicle in question is required. Indeed, if the autonomous vehicle does not detect (or does not detect in an sufficiently early or stable manner) another vehicle in a neighboring lane and changes lanes, a collision between the two vehicles is likely to occur.

[0004] In such a context, most existing vehicle detection and tracking methods rely on classification algorithms that identify vehicles from image sequences or video streams acquired by vision sensors (typically, a camera). Such vision sensors are generally positioned at the front and / or rear of the vehicle in question, so as to acquire images and / or image sequences of a field of view (FoC) in front of and / or behind the vehicle. The classification algorithms can then identify one or more vehicles adjacent to the vehicle in question based on frontal recognition of adjacent vehicles (i.e., by identifying and classifying vehicles detected in the FoC from their front or rear).

[0005] However, such detection and tracking methods using classifiers of Vehicle detection proves ineffective when the front or rear of neighboring vehicles is not yet detectable within the field of view, even if such neighboring vehicles are partially visible, for example, through a partial side view. Thus, if the vehicle in question is in a highway lane and a neighboring vehicle is positioned relatively at the same level as (or slightly behind) the vehicle in question in an adjacent lane, a camera mounted at the rear of the vehicle in question would capture a portion of the rear side of the neighboring vehicle. In such a situation, the front of the neighboring vehicle is not yet visible to the camera mounted at the rear of the vehicle in question, for example, until the speed difference between the vehicle in question and the neighboring vehicle is large enough for the vehicle in question to completely overtake the neighboring vehicle.Until such a complete overtaking maneuver occurs, the adjacent vehicle is therefore at approximately the same level as the vehicle in question, and a lane change by the vehicle in question into the adjacent lane would result in a collision with the adjacent vehicle. Such a critical situation is therefore not detectable by existing classification methods.

[0006] Furthermore, most existing vehicle tracking methods rely on predefined dimensions associated with detected vehicle types. Such dimensions are generally not updated, so even if vehicle tracking is implemented (e.g., by a classifier), such tracking does not provide reliable knowledge of the vehicle's dimensions and therefore does not allow for accurate tracking of the extent of the space occupied by the detected vehicle in the environment of the main vehicle.

[0007] There is therefore a need to secure the decision-making of driver assistance and / or autonomous driving systems, particularly in such a lane change decision context. In particular, there is a need for early and reliable detection, tracking, and awareness of vehicles surrounding a given vehicle even before the front or rear of such surrounding vehicles are visible within the field of view of a vision sensor of the vehicle concerned—and thus before conventional tracking of such surrounding vehicles by existing classifiers is possible. Summary

[0008] The present disclosure improves such a situation.

[0009] A method is proposed for tracking at least one neighboring vehicle present in the environment of a main vehicle, said neighboring vehicle and said main vehicle being motor vehicles, the method being implemented by a device configured to provide a driving assistance function for the main vehicle, said device being connected to at least one camera mounted on the main vehicle and capable of acquiring images of a scene surrounding the main vehicle according to at least one field of view and at acquisition times, the method comprising the following steps: - to detect, from at least one initial image acquired at a first acquisition time, a first object presenting a vertical plane extending laterally relative to the main vehicle, - to determine, according to a predefined coordinate system, at least one piece of data relating to a first position associated with the first object detected, - create a virtual object associated with a set of parameters including at least one current position associated with the virtual object, said current position being estimated from at least said data relating to the first position according to the predefined coordinate system, - to estimate, based at least on the current position of the virtual object, a future position of the virtual object, and - update the parameters associated with the virtual object from at least said future position, said update of the parameters associated with the virtual object including, if a first criterion is satisfied, an update of a current length of the virtual object.

[0010] Consequently, the proposed method allows for the implementation of early tracking of a vehicle adjacent to the main neighbor, namely as soon as such a neighboring vehicle enters, even partially, the field of view of a camera mounted on the main vehicle. In other words, the proposed method makes it possible to track a vehicle adjacent to the main vehicle despite the absence of visibility of a front or rear face of the neighboring vehicle, on which most existing classifiers rely for vehicle tracking.

[0011] Furthermore, the proposed method allows for updating the vehicle length during the tracking process. Thus, unlike most existing classifiers, the proposed tracking method does not rely solely on pre-existing vehicle dimensions, but allows for updating the vehicle dimensions.

[0012] A driving assistance function of the main vehicle refers to a feature of an on-board system on the main vehicle that assists, guides, or even directs the main vehicle's direction. Such a driving assistance function can be implemented in the context of a semi-autonomous or autonomous vehicle, for example, with or without a driver. A driving assistance function can include assistance with driving, lane changes, overtaking another vehicle, or parking, for example.

[0013] A field of view refers to a portion of the main vehicle's environment covered by the camera's field of view. In the context of a setup Depending on the length of the vehicle, the field of view considered is a rear field of view of the main vehicle.

[0014] By images acquired at acquisition times, reference is made to a continuous or discretized succession of images (i.e., in two dimensions) of the portion of the environment contained within the field of view, associated with a succession of instants at which these images were acquired. In particular, the movement of the elements surrounding the main vehicle over time implies a field of view covering an evolving environment: the vehicles and elements neighboring the main vehicle exhibit relative motion with respect to the main vehicle (as the main vehicle moves) and / or absolute motion within the environment.

[0015] A first object refers to an object detected in the field of view. In particular, such an object may be associated with the movement (at least relative to the main vehicle) of a moving (i.e., non-inert) body in the vicinity of the neighboring vehicle. Such a moving body may, for example, be another vehicle near the main vehicle, an obstacle in the environment, or even a bird. Such a first object may, for example, correspond to an optical flow detected in the image, a set of coplanar optical flows or a segment detected in the image, a polygon, or an instance detected in the image (for example, by a classifier). In particular, a first object having a vertical plane extending laterally from the main vehicle refers to an object detected laterally from the main vehicle.In other words, within the field of view, a so-called lateral or oblique view of the object is detected. Such a lateral view is detected, for example, in the case of a neighboring vehicle traveling in a lane adjacent to the main vehicle and downstream of the main vehicle.

[0016] A first position data point associated with the first detected object refers to a measurement and / or estimated or determined value that allows the detected object to be positioned in the acquired image and / or in the environment. Such data may, for example, correspond to a distance between the detected object and the main vehicle, or to two- and / or three-dimensional coordinates. The first position may, for example, refer to the position (in the environment and / or in the image) of a point belonging to the detected object.

[0017] A virtual object is defined as a virtually modeled or created object that can be superimposed on an image acquired by the camera. Such a created virtual object is further associated with the detected object in the sense that the virtual object is estimated to be superimposed on a location presumed to be occupied by the detected neighboring vehicle. Such a virtual object is characterized, in particular, by a position in the environment and / or in the image. Such a virtual object can also be characterized by dimensions so as to reflect presumed dimensions of the detected vehicle. Thus, the virtual object is created to model an occupancy space of the neighboring body (e.g., a neighboring vehicle) detected via the detected object.

[0018] A set of parameters refers to parameter values ​​associated with a virtual object and thus with the virtual object's occupancy space. Such a parameter set may include the virtual object's current position, its current velocity, and / or its current dimensions. This parameter set allows the virtual object to be modeled in the image and / or environment coordinate system. In particular, the current position associated with the virtual object may refer to the position of a reference point associated with the virtual object, such a reference point being, for example, modeled as belonging to the front (and therefore the head) of the detected neighboring vehicle. Such a current position may be imprecise, notably because the front of the detected neighboring vehicle may not be visible in the field of view at the stage of implementation of the proposed method.Thus, the current position can be determined based on one or more assumptions fixing such a current position in the image, for example on an edge of the image.

[0019] A future position is defined as a position estimated or predicted by a data estimator. In particular, the estimation or prediction of such a future position can be implemented based at least on current data fed to the estimator. In one embodiment, the future position of the virtual object is predicted and / or estimated by a Kalman filter from, at least, the current position of the virtual object.

[0020] According to another aspect, a device is proposed that is configured to provide a driving assistance function for a main vehicle, said device being connected to at least one camera mounted on the main vehicle and capable of acquiring images of a scene surrounding the main vehicle according to at least one field of view and at acquisition times, in which the device includes at least one processing circuit configured to implement a method of tracking at least one neighboring vehicle present in an environment of the main vehicle.

[0021] According to another aspect, a computer program is proposed comprising instructions for implementing the tracking method when this program is executed by a processor.

[0022] According to another aspect, a non-transient recording medium is proposed that stores instructions for the implementation of the tracking method when executed, via a program, by at least one processor.

[0023] The features described in the following paragraphs may optionally be implemented independently of each other or in combination. combination with each other:

[0024] In one embodiment, the update of the current length of the virtual object includes extending the length of the virtual object by a value determined from a difference between the current position and the future position, while keeping the current position of the virtual object constant.

[0025] Consequently, updating the parameters allows the difference in position between a current position and a future position predicted from at least one such current position to be projected along the length of the virtual object. In particular, the current position is maintained. Such a tracking method is particularly advantageous, especially when the front of the detected neighboring vehicle is not yet visible in the field of view. Indeed, when a current position of the virtual object is, for example, associated with the position of a front of the virtual object (modeling, for example, the front of the detected neighboring vehicle), maintaining the current position allows for the accurate modeling of situations in which the neighboring vehicle is not yet fully visible (i.e., including its front) in the field of view.Furthermore, once the front of the vehicle is visible and detectable, for example by a classifier, the proposed method advantageously ensures continuous tracking of the neighboring vehicle, with little variation between the location of the virtual object detected by the classifier and the location of the virtual object detected via the proposed method. Moreover, reporting such a difference in position then allows for estimating the length of the detected neighboring vehicle and thus updating the space occupied by the already created virtual object. Therefore, the proposed tracking method allows for updating vehicle tracking parameters that differ from the future position as estimated, for example, by data estimators.

[0026] In one embodiment, the update of the current length of the virtual object is implemented if a first comparison between the future position and the current position indicates a backward position of the virtual object in the first image acquired relative to the current position.

[0027] Therefore, the proposed tracking method allows the length of the virtual object modeling the neighboring vehicle to be updated based on an initial comparison of the virtual object's current and predicted positions. In particular, when it is detected that the virtual object is moving backward in the image (for example, resulting in a lateral component of the future position that is shorter than a lateral component of the current position), a systematic lengthening of the virtual object can be implemented. Specifically, such an update of the virtual object's length can rely solely on current and future parameters associated with the virtual object, without requiring new object detection in the acquired images.

[0028] In one embodiment, said data relating to the first position includes a first lateral distance associated with the first object detected for the first acquisition time, said first lateral distance being determined from measurements of the first object detected in the first image acquired.

[0029] Consequently, the proposed tracking method makes it possible to determine a lateral distance existing between the detected neighboring body, for example, corresponding to the neighboring vehicle, and the main vehicle. The first position, corresponding for example to the position of a specific point on the first detected object, can in particular allow information to be deduced concerning the proximity of the detected neighboring vehicle in the environment relative to the main vehicle.

[0030] A first lateral distance refers to a lateral or oblique separation between the element associated with the first detected object and the primary vehicle. For example, in the case of a primary vehicle and a neighboring vehicle traveling in two parallel lanes (e.g., two highway lanes), the lateral distance corresponds to the lateral separation between the lanes occupied by the primary and neighboring vehicles. Such a lateral distance thus reflects a separation in the environment between the primary and neighboring vehicles.

[0031] In one embodiment, the future position is associated with an object lateral distance from the virtual object.

[0032] Consequently, once the virtual object is created, its movement in the environment can be estimated, for example by a data estimator, so that predicted, or future, parameters such as a future position and lateral object distance can be estimated for such a virtual object.

[0033] In one embodiment, the monitoring method further comprises: - to obtain, from at least one second image acquired at a second acquisition time subsequent to the first acquisition time, data relating to a second position associated with a second detected object, said data relating to the second position including a second lateral distance associated with the second detected object for the second acquisition time, - if the second detected object is determined to belong to the virtual object, compare, in a second comparison, the second lateral distance and the object's lateral distance, in which the update of the parameters associated with the virtual object also depends on a result of the second comparison.

[0034] Consequently, the proposed tracking method makes it possible to take into account changes in the occupancy of the neighboring vehicle detected in the environment. Thus, in a second image acquired at a later acquisition time than the first acquisition time, the second detected object may differ from the first detected object and allows for updating the changes in the space occupied by the neighboring vehicle. in the environment (for example, the neighboring vehicle is increasingly entering the field of view, or the speed difference between the main and neighboring vehicles is such that the main vehicle has completely overtaken the neighboring vehicle and its front end is now visible in the field of view). The proposed update therefore allows, based on a second criterion distinct from the first, the virtual object to be updated.

[0035] A second detected object refers to a distinct object, whether or not it is separate from the first object detected in an acquired image. Such a second detected object may then correspond to an instance, a polygon, an optical flow, or a segment reflecting a movement of the neighboring vehicle detected in the environment. The second position associated with such a second detected object may then reflect a new space occupied by the neighboring vehicle.

[0036] In one embodiment, if the second lateral distance is strictly less than the object lateral distance, the update of the parameters associated with the virtual object includes an update of the current position of the virtual object, said update of the current position including at least an update of a lateral component of the current position of the virtual object corresponding to a repositioning of the virtual object.

[0037] Consequently, the length update is contingent upon a comparison of lateral distances from successively detected objects. In particular, if the second lateral distance is greater than or equal to the estimated lateral distance of the virtual object, an update of the virtual object's current position is not required, as the neighboring vehicle is considered to be gradually entering the field of view. Thus, the proposed tracking method takes advantage of the potential absence of a visible front end in the field of view to maintain the vehicle's current position (and therefore avoid modeling a virtual object "receding" in the acquired image) and compensates for the difference between the current and future positions by laterally lengthening the neighboring vehicle.

[0038] In particular, the method advantageously corrects potential overestimations of the determined and / or predicted lateral distances between the neighboring vehicle and the main vehicle. Such a correction allows the tracking method to gain in accuracy and avoid dangerous situations in which the neighboring vehicle would actually be closer to the neighboring vehicle than estimated (the second lateral distance corresponding to the second object detected at a more recent time is shorter than the estimated object lateral distance).

[0039] Consequently, the proposed tracking method makes it possible to take into account inaccuracies related to the creation of the virtual object based on the parameters and the current position. Indeed, if the second lateral distance is strictly less than The lateral object distance, the virtual object estimated based on the lateral object distance, does not reflect the space actually occupied by the neighboring vehicle, and in particular, the proximity of the neighboring vehicle to the main vehicle has been underestimated. The proposed method corrects this potentially critical situation (since the neighboring vehicle is potentially closer to the main vehicle than the location modeled by the virtual object via its current position) by updating the lateral position of the virtual object to reflect the actual lateral proximity between the neighboring and main vehicles. Thus, the method advantageously maintains consistency between the update of the virtual object and what is detectable in the second image.

[0040] In one embodiment, if the second lateral distance is strictly less than the object's lateral distance, the update of the parameters associated with the virtual object includes an update of the virtual object's current position, said update of the current position further including an update of a height component of the virtual object's current position corresponding to a repositioning of the virtual object. Such an update of the height component may further be based on terrain topology information in the environment.

[0041] In one embodiment, the current position of the virtual object is associated with the position of a first reference point belonging to the virtual object, said position of a first reference point belonging to the virtual object being determined such that, in the predefined coordinate system corresponding to an image frame in the acquired images, the first reference point is positioned on a vertical end edge of the first image.

[0042] Consequently, the current position allows the location of the neighboring vehicle to be modeled in the environment despite a partial side view of the detected vehicle. The current position then obeys constraints by modeling the location of the neighboring vehicle on the vertical edge of the image.

[0043] A first reference point refers to a point estimated to belong to the neighboring vehicle detected in the environment. In particular, such a first reference point can model the location of a front face of the neighboring vehicle (which may not yet be visible in the field of view).

[0044] In one embodiment, the virtual object is created from the position of the virtual object and predefined dimensions associated with a type of motor vehicle, at least one predefined dimension corresponding to the length of the virtual object.

[0045] Consequently, the proposed tracking method makes it possible to model, via the virtual object, a space occupied by the neighboring vehicle, despite the absence (at least during an initial iteration of the tracking method) of information relating to the dimensions of the detected vehicle.

[0046] By the length of the virtual object, reference is made to a lateral length of the virtual object, such a virtual object modeling a neighboring vehicle from an object detected at an angle (along a lateral plane) relative to the main vehicle.

[0047] In one embodiment, the future position of the virtual object is estimated from a plurality of positions of the virtual object respectively determined for the acquisition times until at least one frontal portion of the neighboring vehicle is detected in the field of view of the camera on at least one of the acquired images.

[0048] Consequently, the proposed tracking method can advantageously be implemented so that before a front (or frontal) face of the neighboring vehicle is visible, thus enabling early tracking of vehicles adjacent to the main vehicle. Therefore, when a front (or frontal portion) of the neighboring vehicle becomes visible in the field of view, the existing classifiers can take over vehicle tracking by updating the virtual object.

[0049] In one embodiment, each of the positions of the plurality of positions of the virtual object is associated with an uncertainty value relative to an uncertainty on the initial conditions, said uncertainty value being greater than a predefined uncertainty threshold.

[0050] Consequently, the parameters determined to model the virtual object based on a biased view of the detected neighboring vehicle can be assigned an uncertainty value higher than a threshold value, corresponding, for example, to an uncertainty value of data determined by a conventional classifier. Thus, when the parameters determined within the framework of the proposed method are fed into a data estimator for predicting a future position, a weight relative to a degree of error is assigned to the parameters. Such an uncertainty value can, for example, correspond to a variance or a covariance. Thus, when the estimator predicting the future position receives other parameter values ​​exhibiting a lower degree of uncertainty, the estimator can predict the future position of the virtual object based on these values.For example, when the front of a neighboring vehicle becomes visible in the field of view, a vehicle front face classifier can then provide the estimator with more reliable data than data obtained from a position estimation based on a detected lateral object. The proposed method thus allows for a transition between different data sources, enabling the tracking of the neighboring vehicle.

[0051] In one embodiment, the parameters associated with the virtual object are updated by keeping a parameter of the virtual object constant.

[0052] Consequently, the proposed method makes it possible to maintain the stability of the virtual object during the updating of the parameters ensuring the tracking of the neighboring vehicle. Indeed, the proposed method makes it possible to update both a length and / or a position of the virtual object. A parameter associated with the virtual object is then kept constant, for example at least one component of the position associated with the virtual object, when the length of the virtual object is updated, or a temporal parameter associated with the virtual object such as a collision time or the product of the velocity by a distance of the virtual object, when the position of the virtual object is updated. Brief description of the drawings

[0053] Other features, details and advantages will become apparent upon reading the detailed description below, and upon analysis of the accompanying drawings, on which:

[0054] [Fig-1] Fig. 1 shows a schematic representation of a main vehicle according to a mode of realization.

[0055] [Fig.2] Fig.2 shows an aerial view of a scene surrounding the vehicle principal according to a mode of embodiment.

[0056] [Fig.3] Fig.3 shows a schematic representation of a device for assisting the conducted according to a method of implementation.

[0057] [Fig. 4] [Fig. 4] shows steps in a method of tracking a vehicle according to a method of implementation.

[0058] [Fig. 5] [Fig. 5] shows a photograph of a scene surrounding the vehicle principal to an initial acquisition phase according to a method of implementation.

[0059] [Fig.6] Fig.6 shows a virtual tracking object created according to one embodiment.

[0060] [Fig.7] Fig.7 shows an update of the virtual tracking object created according to a method of implementation.

[0061] [Fig.8] Fig.8 shows a photograph of a scene surrounding the vehicle main to a second acquisition time according to a mode of realization.

[0062] [Fig.9] Fig.9 shows an update of the virtual tracking object created according to a method of implementation.

[0063] [Fig. 10] The [Fig. 10] shows a shot of a scene surrounding the main vehicle at a third acquisition time according to one embodiment.

[0064] [Fig. 11] The [Fig. 11] shows a measurement taken on a shot of a scene surrounding the main vehicle at a first acquisition time according to an embodiment.

[0065] [Fig. 12] The [Fig. 12] shows a reference point associated with a neighboring vehicle according to one embodiment. Description of the embodiments

[0066] Reference is made to [Fig. 1]. [Fig. 1] schematically represents a principal vehicle VP. The principal vehicle VP may be a motor vehicle. There is no limitation on the type of vehicle to which the principal vehicle VP belongs. The principal vehicle VP may, for example, be a passenger car, a commercial vehicle, an industrial vehicle, and may For example, it could be a car, van, motorcycle, truck, or bus. The main vehicle (MV) can also be a towed vehicle, such as one pulling a trailer, semi-trailer, or caravan. The dimensions of the main vehicle (MV) can range from 2 meters to 20 meters in length, from 0.5 meters to 5 meters in width, and from 1 meter to 5 meters in height.

[0067] The main vehicle VP is equipped with at least one on-board system providing a plurality of functions or applications of the main vehicle VP. Such functions may, for example, correspond to cruise control, power steering, automated airbag deployment, automatic headlight adjustment...

[0068] The main vehicle VP is specifically equipped with an on-board system enabling the detection of objects surrounding the main vehicle VP, for example, as part of a driver assistance function such as obstacle detection, lane change assist, or automatic lane change. To this end, the on-board system of the main vehicle VP includes a driver assistance device 2. Device 2 is capable of providing a driver assistance function based on multiple views (or images) of the environment surrounding the main vehicle VP. For this purpose, Device 2 is notably connected to one or more vision sensors, such as a camera or a camera 1 (the vision sensor will be considered a camera 1 in the remainder of this description). Device 2 may also be capable of transmitting or communicating data, for example, data relating to the driver assistance function being performed.For this purpose, device 2 can be connected to a communication interface 50. In a particular embodiment, the interface 50 can be integrated into device 2. Such a communication interface 50 can, for example, be a human-machine interface. The interface 50 can include a display screen, a touchscreen, a dashboard, and / or a loudspeaker. The data transmitted by device 2 to the interface 50 can, for example, correspond to assistance information indicating to the driver of the main vehicle VP whether or not they can change lanes.

[0069] Reference is now made to [Fig. 2]. [Fig. 2] illustrates an environment (or scene) ENV, in which the main vehicle VP is located. [Fig. 2] is an aerial view of such an ENV scene.

[0070] The ENV environment can be defined in a three-dimensional frame (X,Y,Z), referred to as a "world frame" as illustrated in Figures 1 and 2. The origin of such a world frame (X,Y,Z) is predefined and fixed in the ENV environment.

[0071] The main vehicle VP is considered to be in motion in the ENV scene. Such an ENV scene corresponds, for example, to a road or highway consisting of several traffic lanes. These traffic lanes may, in particular, be pa parallel to each other, as represented by the vertical dashed lines in [Fig. 2]. [Fig. 2] illustrates, for example, three traffic lanes, with the main vehicle VP located in the middle lane. The main vehicle VP is considered to be moving along the principal direction X of the (X,Y,Z) coordinate system, such a principal direction being called the longitudinal direction. This movement is shown schematically in Figures 1 and 2 by an arrow attached to the main vehicle VP. The main vehicle VP can also move along a direction Y of the (X,Y,Z) coordinate system, called the lateral direction. Such movement along the Y direction, called lateral movement (or displacement), can occur, for example, when the main vehicle VP changes lanes. A displacement along the Z direction, that is, vertically, of the main vehicle VP is considered absent or negligible.The main vehicle VP is therefore considered to be fixed to the ground and thus has a constant height along the Z direction, corresponding to a predefined dimension of the main vehicle VP. In one embodiment, such a height along the Z direction of the main vehicle VP may vary on the order of centimeters or decimeters, such a variation in height along the Z direction being, for example, linked to shock absorbers of the main vehicle VP and / or to reliefs or asperities present in the ENV environment (particularly on the traffic lane of the main vehicle VP). In the remainder of the description, such a height along the Z direction of the main vehicle VP is considered to be known.

[0072] The ENV scene surrounding the main vehicle VP also includes other OV elements and vehicles VV1, VV2, VV3. Vehicles VV1, VV2, VV3 are neighboring vehicles of the main vehicle VP. Such neighboring vehicles VV1, VV2, VV3 are, like the main vehicle VP, motor vehicles moving within the ENV scene. Such neighboring vehicles VV1, VV2, VV3 may, in particular, have dimensions and movement characteristics (in terms of speed or acceleration, for example) similar to or different from the main vehicle VP. For example, [Fig. 2] may represent a main vehicle VP moving on a highway lane and neighboring vehicles VV1, VV2, VV3 traveling on highway lanes adjacent to the lane used by the main vehicle VP. OV objects are neighboring elements of the main vehicle VP and can refer to any distinct element of a neighboring vehicle VV1, VV2, VV3.A neighboring element OV can, for example, correspond to an obstacle located in the scene such as a central reservation separating two traffic lanes, a signpost or even a bird flying in the ENV scene.

[0073] As illustrated in [Fig. 2], the main vehicle VP is considered to be equipped with a camera 1 positioned at the rear of the main vehicle VP. In another embodiment (not shown in [Fig. 2]), the camera 1 may be positioned at the front of the main vehicle VP or several cameras 1 can be positioned both at the front and rear of the main vehicle VP. In the context of this description, camera 1 is considered to be positioned at the rear of the main vehicle VP and the field of view (FOV) is a rear field of view of the main vehicle VP. Camera 1 captures images (or shots) of the ENV scene of the main vehicle VP within a field of view (FOV). Such a field of view depends, among other things, on the type of camera 1 used and the positioning of camera 1 in (or on) the main vehicle VP. With reference to [Fig. 2], the field of view (FOV) covers a portion of the ENV scene, so that a limited portion of the ENV scene is captured within the field of view (FOV). Thus, in [Fig. 2], the camera 1 is positioned at the rear of the main vehicle VP.[2] The field of view (FOV) shown covers a portion of the traffic lane in which the main vehicle (VP) is located and respective portions of the adjacent traffic lanes. In particular, the FOV covers a portion of the scene within the environment (ENV) in which the adjacent vehicle (VV2) is located entirely, and a portion of the scene within the environment (ENV) in which the adjacent vehicle (VV1) is partially located (unhatched gray area). For example, as illustrated in [Fig. 2], the left rear end of the adjacent vehicle (VV1) (e.g., including the left rear wheel of the adjacent vehicle (VV2)) is within the FOV. However, a reference point (W), for example, located at the midpoint of the front of the vehicle, is not within the FOV in [Fig. 2]. The adjacent vehicle (VV3) is not visible within the FOV of the main vehicle (VP). A portion of the adjacent vehicle (VV1) is not visible within the FOV of the main vehicle (VP).The parts of neighboring vehicles VV1 and VV3 not visible in the field of view (FOV) are represented in [Fig. 2] by striped areas. A neighboring object OV, for example a bird, may be visible in the FOV.

[0074] In the context of an ENV scene as represented in Figure 2 with respect to the coordinate system (X, Y, Z), it is assumed, for example, that the main vehicle VP is moving at a known main speed y^. To facilitate the following description, such a main speed y^ can be considered to have a constant magnitude VVp. The neighboring vehicle VV1 is considered to be moving at a neighboring speed y^ unknown to the driving assistance device 2 of the main vehicle VP. In the embodiment described below, it can be assumed that the magnitude VVp of the neighboring speed y^ is less than the magnitude VVp of the main speed y^.Under such an assumption and under the assumption that the neighbor speed y^ of the neighbor vehicle VV1 remains substantially constant over a considered time interval, like the main speed y^ of the main vehicle VP, the difference in distance along the longitudinal direction X between the . The main vehicle VP and the neighbor vehicle VV1 will increase over time, so that at a future time, the overtaking of the main vehicle VP on the neighbor vehicle VV1 will be such that the neighbor vehicle VV1 will be entirely included in the (rear) field of view (FOV) of the main vehicle VP, as is the initial case of the neighbor vehicle VV2.

[0075] Such an evolution of the belonging of the neighbouring vehicle VV1 to the FOV field of view of the main vehicle VP over time is illustrated in figures 5, 8 and 10.

[0076] Reference is made to Figures 5, 8, and 10. Figures 5, 8, and 10 schematically represent views of the ENV scene surrounding the main vehicle VP according to the FOV of camera 1 of the main vehicle VP, as positioned in [Fig. 2]. To facilitate the readability of Figures 5, 8, and 10, only the portion of the neighboring vehicle VV1 visible in the FOV is shown in Figures 5, 8, and 10; the neighboring vehicle VV2 and the neighboring element OV of [Fig. 2] included in the FOV are not shown in the views in Figures 5, 8, and 10.

[0077] Figures 5, 8, and 10 can correspond to images successively acquired by camera 1 at successive instants (or acquisition times) T1, T2, and T3. Each of the instants T1, T2, and T3 can, for example, be separated by one or more milliseconds in time. Each image corresponds to a set of pixels with definable coordinates in a two-dimensional coordinate system (y,z), called the "image coordinate system," as shown in Figures 5, 8, and 10. For example, the lowest and leftmost pixel of each image acquired by camera 1 can correspond to the origin of the image coordinate system (y,z). In other embodiments, the central pixel of the image acquired by camera 1, the optical center of the image, the pixel of the image associated with the location of camera 1 in the ENV environment or the pixel corresponding to a vanishing point C' of the acquired image can be considered as the origin of the (y,z) coordinate system.The resolution of the acquired images (and therefore the pixels) is the same for all acquired images and depends in particular on the properties of the camera 1.

[0078] Figure 5 schematically represents an image acquired by camera 1 at time T1, corresponding, for example, to a configuration of the scene ENV and the field of view FOV illustrated in Figure 2. Thus, only a first, so-called lateral portion of the neighboring vehicle VV1 is visible in the image of Figure 5; this first lateral portion includes, in particular, the left rear wheel of the neighboring vehicle VV1.

[0079] Figure 8 schematically represents an image acquired by camera 1 at time T2 following time TL. At such time T2, a second lateral portion of the neighboring vehicle VV1, larger than the first lateral portion, is visible in the image of Figure 8. This second lateral portion always includes, in particular, the left rear wheel and partially the left front wheel of the neighboring vehicle VV1. In other words, the difference The speed differential between the main vehicle VP and the neighboring vehicle VV1 between times T1 and T2 is such that the neighboring vehicle VV1 "enters more and more" into the field of view FOV of camera 1 between times T1 and T2.

[0080] Figure 10 schematically represents an image acquired by camera 1 at time T3 following time T2. At such time T3, the entirety of the neighboring vehicle VV1 is visible in the image of Figure 10. In particular, a reference point W, shown in Figure 2 and positioned on the front face of the neighboring vehicle VV1, has become visible in the acquired image; this reference point W corresponds to pixel w in the image of Figure 10. In other words, the speed differential between the main vehicle VP and the neighboring vehicle VV1 between times T2 and T3 is such that the neighboring vehicle VV1 has "entirely entered" the field of view (FOV) of camera 1 between times T2 and T3.

[0081] In the context of a driver assistance function provided by device 2, aimed for example at assisting the main vehicle VP in a lane change maneuver, a process for detecting and tracking vehicles surrounding the main vehicle VP is required, so that the main vehicle VP does not move into a neighboring traffic lane if it risks colliding with one of the surrounding vehicles. Existing techniques for detecting and tracking vehicles surrounding the main vehicle VP rely in particular on the use of classifiers, based for example on convolutional neural networks (CNNs), K-nearest neighbors (KNNs), or support vector machines (SVMs).Such classifiers, enabling the detection and tracking of vehicles surrounding the main vehicle VP, rely in particular on the detection and classification of a frontal (front or rear) view of the vehicles surrounding the main vehicle VP. The use of such classifiers may include a learning phase based on a plurality of images representing frontal views of various types of vehicles. Thus, with reference to [Fig. 2], a frontal view of the neighboring vehicle VV2 belonging to the FOV of camera 1, the detection and tracking of the neighboring vehicle VV2 can be implemented based on existing classification techniques. The same applies to the neighboring vehicle VV1 in the image of [Fig. 10] at time T3.

[0082] However, the positions of the neighboring vehicle VV1 shown in [Fig. 2] and Figures 5 and 8 correspond to situations in which existing classifiers are unable to detect or track the neighboring vehicle VV1 effectively, or at least not without generating significant costs and / or computation time, due to the absence of a visible frontal view of the neighboring vehicle VV1 in the photographs of the Figures 5 and 8 (the reference point W of [Fig. 2] positioned on the front face of the neighboring vehicle VV1 is not yet visible in the FOV of camera 1 at these stages). These situations nevertheless correspond to critical situations in which the main vehicle VP could collide with the neighboring vehicle VV1 if the main vehicle VP changes position and moves into the lane of the neighboring vehicle VV1, in the absence of lane change assistance information indicating that the neighboring vehicle VV1 is detected near such a position.

[0083] A method for anticipating the tracking of the neighboring vehicle VV 1 is then proposed and detailed in [Fig. 4] to detect and track the neighboring vehicle VV1, particularly at the stage of the situations represented in Figures 2, 5, and 8. Such a stage is said to be "anticipated" in that a frontal face of the followed vehicle may not yet be visible in the field of view (FOV). Such an anticipatory tracking method for the neighboring vehicle VV 1 is implemented by a driver assistance system shown in [Fig. 3].

[0084] Reference is now made to [Fig. 3]. [Fig. 3] represents a schematic representation of an embedded system of a main vehicle VP. In particular, such an embedded system corresponds to a driver assistance system for the main vehicle VP. The driver assistance system of the main vehicle VP notably provides a lane change assistance or automatic lane change function for the main vehicle VP when the main vehicle VP is moving in a traffic lane as shown, for example, in [Fig. 2].

[0085] The system first includes the driving assistance device 2. The device 2 itself includes a unit 20 for detecting objects on acquired images, a unit 30 for determining parameters associated with the objects detected by the unit 20, a unit 40 for creating virtual objects associated with the detected objects, a unit 50 for estimating data, such as a Kalman filter, (for example by fusion, by learning and / or by prediction of data) for tracking created virtual objects and a unit 60 for updating parameters associated with the created virtual object.

[0086] The driving assistance device 2 is further connected, on the one hand, to a vision sensor of the type camera or camera 1. The device 2 may also include an input unit (not shown in [Fig.3]) enabling the device 2 to receive in substantially real time a data stream from the camera 1. Such a data stream corresponds to a discrete or continuous succession of images (or shots) of the ENV scene according to the field of view FOV of the camera 1. Each received image is associated with an image acquisition time by the camera 1. Each image can be time-stamped.

[0087] On the other hand, the driver assistance device 2 can be connected to a communication interface 70. Such an interface 70 can correspond to a human-machine interface integrated into the on-board system of the main vehicle VP. Such an interface 70 can also be integrated into device 2. The communication interface 70 can also be a remote interface. The communication interface 70 can include a display screen, a touch screen, a dashboard, or a loudspeaker, enabling the transmission, for example, by visual, haptic, and / or audible information, of indications relating to driver assistance for the main vehicle VP. In particular, the communication interface allows the transmission of a detection status of a vehicle adjacent to the main vehicle VP or an indication relating to a lane change possibility for the main vehicle VP. Such information transmitted by the interface 70 can, for example, correspond to an estimated aerial visual representation of the respective positions of the main vehicle VP and the adjacent elements belonging to the FOV in real time (e.g., as shown in [Fig.2]), an audio stimulus alerting of a risk of collision, or a superimposition of a virtual object OV1 highlighted and updated in real time on a stream of images from camera 1, as illustrated for example in figures 6, 7 and 9. .

[0088] The units 20, 30, 40, 50, and 60 of the device 2 each comprise a processing circuit including at least one processor (21, 31, 41, 51, 61) and one memory unit (22, 32, 42, 52, 62) in order to implement one or more steps of the method for anticipating the tracking of a neighboring vehicle VV1, which will be described in [Fig. 4]. In particular, each processing unit 20, 30, 40, 50, and 60 of the device 2 can rely on data processed and / or obtained by other units in order to implement one or more steps of the method for anticipating the tracking of a neighboring vehicle VV1, as illustrated by the arrows in [Fig. 3].

[0089] Reference is now made to [Fig. 4]. [Fig. 4] illustrates a series of steps in the implementation of a method for the early tracking of a neighboring vehicle VV1 by a system including a driving assistance device 2 for a main vehicle VP as shown in [Fig. 3]. In particular, the method for the early tracking of a neighboring vehicle VV1 described in [Fig. 4] comprises a phase of detecting an object associated with the neighboring vehicle VV1 and a phase of tracking, strictly speaking, the neighboring vehicle VV1 from the detected object.

[0090] At a step 400, the device 2 receives a plurality of images (or shots) associated with respective image acquisition times from the camera 1. Such images may, in particular, be received continuously, for example via a video stream. In such a case, at step 400, the device 2 can discretize the received video stream so as to obtain a set of discrete time-stamped images associated with respective acquisition times.

[0091] Steps 410 to 450 described below are implemented starting from an object detected on an acquired image - for example a first detected object OBJ1 - to a At a given acquisition time – for example, the first acquisition time Tl

[0092] – a first object OBJ1 is detected in at least one of the acquired images. With reference to [Fig. 5], such a first detected object OBJ1 is represented in the image corresponding to the first acquisition time Tl. In one embodiment, the first detected object OBJ1 is a so-called lateral object in the acquired image, in that the first detected object OBJ1 belongs to a lateral plane of the image, such a lateral plane being, for example, parallel to the (X,Z) plane in the world frame (X,Y,Z). The detection of such a first lateral object OBJ1 is then said to be anticipated in that it occurs before a front face of the neighboring vehicle VV1 is visible. In particular, in step 410, the first detected object OBJ1 corresponds to a non-inert body.In other words, the first detected object OBJ1 corresponds to a body moving in the environment ENV (and therefore, in the world frame (X,Y,Z)) and exhibiting a velocity (for example, a so-called longitudinal velocity corresponding to a displacement along the X axis in the world frame (X,Y,Z) and / or a so-called lateral velocity corresponding to a displacement along the Y axis and / or a rotational velocity).

[0093] The first object OBJ 1 can be detected in step 410 on a first image (for example, the image in [Fig. 5]) by processing one or more successively acquired images. In one embodiment, the first object OBJ 1 can, for example, be detected by a classifier capable of detecting a lateral object OBJ 1 on the acquired image, for example, a classifier detecting a lateral portion of a vehicle, such as a vehicle wheel detector. Alternatively, the first object OBJ 1 can be detected by homography on at least two successively acquired images, so as to identify one or more connected sets, or optical flows, of relatively homogeneous velocities on the images within a defined time interval. According to such an embodiment, the detection of the first object OBJ 1 then includes determining a pixel displacement at a substantially common velocity from one image to the next, or optical flow.Each optical flow can be determined by point matching between several successive images, for example by the Lucas-Kanade method. The determined optical flows can then be segmented into one or more segments having a common point or vanishing line and belonging to lateral planes in the field of view (FOV). In other words, the acquired image can be segmented into one or more segments, each segment defining a grouping of optical flows (and therefore a grouping of pixels having approximately the same speed and describing the same motion within a given time interval).

[0094] Thus, in step 410, the first detected object OBJ1 is a set of pixels that can be formed by an instance, a contour, a polygon, or even a rectangle (or, in English, a "bounding box") detected and classified by a classifier by object segmentation in the acquired image, for example. Alternatively, the first detected object OBJ1 is a set of pixels that can be formed by an optical flow or a segment including several optical flows. In particular, the set of pixels forming the first detected object OBJ1 can be a set of coplanar pixels parallel to a lateral plane in the world coordinate system (X,Y,Z).

[0095] At a step 420, a data point relating to a first position of the first detected object OBJ1 can be determined by the device 2. In particular, such a data point relating to the first position of the first object OBJ1 can be a first lateral distance dkt i associated with the first detected object OBJ1. Such a first lateral distance diat4 can notably be determined from values, measurements, estimates, or data associated with the first detected object OBJ1 on the first acquired image. In one embodiment, such a first lateral distance diat>i can, for example, be a value fixed approximately given an area of ​​the acquired image in which the first detected object OBJ1 is located.In one embodiment, such a first lateral distance diatji can be determined from the number of ground lines visible on the first acquired image between a central vertical line VL of the image (for example, the dashed vertical line) and a ground line at the level of the first detected object OBJ1, for example, in the case of vehicles on highway lanes. In another embodiment, the first lateral distance diat,i can be determined from data associated with the first detected object OBJ1 and / or measured on the first acquired image, such as the size (in pixels) of the first detected object OBJ1 and / or the angle between a pixel belonging to the first detected object OBJ1 and a predefined reference pixel on the first acquired image. Such data can be estimated by a classifier that detected the first object OBJ1.Such data can also be estimated from measurements taken on the first acquired image. For example, the first lateral distance diat4 can be determined from the estimation of a first position associated with the detected object OBJ1. Such a first position can, for example, be the position of a first reference pixel bl belonging to the detected object OBJ1. Such a first reference pixel bl is, for example, shown in [Fig. 5]. In one embodiment, such a first reference pixel bl can be identified by the object classifier as, for example, the wheel of the vehicle. Alternatively, the first reference pixel bl can be identified by measuring a minimum angle 0min, as shown in [Fig. 11]. To measure such a minimum angle 0min, a vanishing point C' of the first acquired image is considered, such a vanishing point coinciding with the optical center of the first acquired image in [Fig. 11].l 1], as well as a vertical axis VL passing through such a vanishing point C'. The first reference pixel bl can then be selected from the set of pixels forming the detected object OBJ1, so that the angle formed between on one side, the vertical axis VL and on the other. part, the axis formed between the vanishing point C' and such a reference pixel bl, i.e. minimal 0min.

[0096] The first reference pixel bl thus obtained then has a first position in the first acquired image, which can be determined by the coordinates of point b1 in the image frame (y,z). From such coordinates in the image frame (y,z), the first position associated with the detected object OBJ1 can be estimated in the world frame (X,Y,Z). In particular, the coordinates of the first position in the world frame (X,Y,Z) can be determined, from the coordinates of the first reference pixel bl in the image frame (y,z) and the predefined characteristics of camera 1, based on the pinhole model. The application of the pinhole model to obtain the coordinates associated with the first detected object OBJ1 in the world frame (X,Y,Z) relies in particular on several computational assumptions related to the properties of camera 1 as well as to the ENV environment associated with the world frame (X,Y,Z).For this purpose, classic extrinsic parameters of camera 1 can be considered. A fisheye camera 1 can be used. Furthermore, a fiat-world assumption can be applied. Alternatively, georeferencing of the ENV environment using ground markers or other topographic surveys can also be considered. In the context of the applied pinhole model, any geometric distortions induced by the optical system of camera 1 can be neglected. In another embodiment, determining the coordinates associated with the first detected object OBJ1 in the world coordinate system (X,Y,Z) can include correcting for the distortion of the camera 1 lens. The first lateral distance diat4 associated with the first detected object OBJ1 (actual, expressed in meters) can then be directly deduced from the Y-coordinate of the first position in the world coordinate system (X,Y,Z).The first lateral distance dkt i associated with the first detected object OBJ1 can be obtained by calculating the difference between the Y-axis coordinate of the first position in the world coordinate system (X,Y,Z) and the Y-axis coordinate of an origin point of camera 1 of the main vehicle VP, for example coinciding with the vanishing point C', the optical center of the image, or a point on the image associated with the location of camera 1 in the ENV environment. Alternatively, a combination of the aforementioned methods can be implemented in step 420 to determine the first lateral distance diat^ associated with the first detected object OBJ1.

[0097] At the end of step 420, the first object detected OBJ1 on the first image acquired at the first acquisition time T1 can then be associated with data relating to a first position of the first object OBJ1. In particular, such data can be a first lateral distance diat1. Such a first lateral distance diat1 corresponds in particular to an estimate of a lateral gap separating the main vehicle VP from The detected object 0BJ1. In other words, step 420 allows us to estimate a real lateral deviation (i.e., within the ENV environment) between the main vehicle VP and a neighboring moving body corresponding to the first detected object OBJ1 (such a neighboring body being, for example, a neighboring vehicle VV1 of the main vehicle VP). In the remainder of this description, it is assumed that such a neighboring moving body corresponding to the first detected object OBJ1 is a neighboring vehicle VV1.

[0098] At a step 430, a set of parameters associated with the neighboring vehicle VV1 thus detected can be determined and / or estimated by the device 2. Such parameters can, in particular, be estimated from the first lateral distance diat>ias associated with the first detected object OBJ1 and corresponding to the neighboring vehicle VV1. Such parameters can, for example, include a current position Pn associated with the detected neighboring vehicle VV1, a current velocity associated with the detected neighboring vehicle VV1, and / or current dimensions sn associated with the neighboring vehicle VV1.

[0099] Thus, in step 430, determining a set of parameters associated with the neighboring vehicle VV1 may include determining a current velocity associated with the neighboring vehicle VV1. In one embodiment, such a current velocity may, for example, be a rotational velocity estimated from the observation of the evolution of a point located on the wheel by the wheel classifier. Alternatively, such a current velocity may, for example, be deduced from the first lateral distance diat,i associated with the first detected object OBJ1 as determined in step 420, as well as from a time datum iTTC associated with the first detected object OBJ1. Such a time datum iTTC may be relative to an inverse quantity of a collision time associated with the first detected object OBJ1. In the case of a first detected object OBJ1 corresponding to a segment (i(to a set of optical flows), such iTTC time data associated with each segment can, for example, be determined by one of the methods described in documents WO 2018059629, WO 2018059631 and WO 2018059632.

[0100] Thus, at step 430, a current velocity vn associated with the neighboring vehicle VV1 can be determined as being a velocity associated with the first detected object OBJ1: XÎTTCoù; - vn is a relative speed (with respect to the main vehicle VP) associated with the first detected object OBJ1 (expressed in meters per second), ■ dlatl is the lateral distance associated with the first detected object OBJ1 (expressed in meters), and - iTTC is the time data associated with the first detected object OBJ 1 (expressed as the inverse of a time).

[0101] Furthermore, in step 430, the determination of a set of parameters associated with the The neighboring vehicle VV1 may include determining current dimensions sn associated with the neighboring vehicle VV1. In one embodiment, such current dimensions s' may be predefined by pre-existing dimensions commonly observed in vehicles. Such pre-existing dimensions may, for example, correspond to a predefined length L1, a predefined width L2, and a predefined height H.

[0102] Finally, in step 430, determining a set of parameters associated with the neighboring vehicle VV1 may include determining a current position Pn associated with the neighboring vehicle VV1. In particular, such a current position Pn may be associated with an estimated point on the neighboring vehicle VV1, for example, a first reference point Wi associated with the neighboring vehicle VV1. Such a first reference point Wi may, in particular, be associated with the center of a front face (not visible in the acquired image of [Fig. 5]) of the neighboring vehicle VV1. Thus, estimating the current position Pn associated with the neighboring vehicle VV1 may amount to estimating the coordinates (X_Wi, Y_Wi, Z_Wi) of the first reference point Wi in the world frame (X, Y, Z). In other words, in one embodiment: P„(X)=X_W1 p,((y)=Y_w], P„(Z)= z_w,

[0103] To estimate the coordinates (X_Wi, Y_Wi, Z_Wi) of the first reference point Wi, in an embodiment represented by [Fig. 12], the device 2 can rely on the coordinates (X_C, Y_C) of a central point C of the camera 1 in the world frame (X, Y, Z), such a central point C corresponding, for example, to the optical center or to a point associated with the location of the camera 1 or of a sensor of the camera 1 in the environment ENV (the projection of such a central point C resulting in point C' on [Fig. 5]). The device 2 can also rely on the pre-existing dimensions L1, L, H of the neighboring vehicle VV1, on the first lateral distance d]ati determined in step 420, on the focal length / of the camera 1 and on the size of the sensor sz_capt in particular. Device 2 can then determine, by applying the pinhole model and Thales' theorem: r_% = y_c±(Kd+H2) x w =x c . mmoù: - । - - (X _ Wp Y _ IVp are the coordinates of the first reference point Wi in the world frame (X,Y,Z), - (X_C, Y_C) are the coordinates of the central point C of camera 1 in the world coordinate system (X,Y,Z), ■ d / at l cst 'a first lateral distance associated with the first detected object OBJ1, - f is the component along the X axis of the focal length of camera 1, _ sz_capt CS( |a actual size of the sensor, and - L2 is the predefined width of the neighboring vehicle VV1.

[0104] Such coordinates in the world frame (X,Y,Z) of the first reference point Wi can be determined, in particular, by considering that the central point C coincides with the center of the acquired image. Optical distortions can be considered negligible. Furthermore, the focal length θ is considered identical on both axes—vertical VL and horizontal—of the image. In particular, the aforementioned formulas describing the coordinates of the first reference point Wi depend on the relative positioning of the neighboring vehicle VV1 with respect to the main vehicle VP.

[0105] Finally, a coordinate ZW of the first reference point Wi along the Z-axis in the world frame (X,Y,Z) can be determined as corresponding to half the predefined height of the neighboring vehicle VV1, namely ZW = H / 2. Alternatively, the coordinate ZW can correspond to the coordinate along the Z-axis of the first position associated with the first detected object OBJ1. In particular, the coordinate ZW can correspond to the height in the world frame (X,Y,Z) of the point closest to the ground whose projection in the image frame (y,z) belongs to the first detected object OBJ1.

[0106] Thus, at the end of step 430, the device 2 can obtain a set of (current) parameters (Pn, V, A) associated with the neighboring vehicle VV1 (and therefore with the first detected object OBJ1). In particular, the parameters (Pn, V) are associated with a given acquisition time—namely, the first acquisition time T1 here—corresponding to a certain positioning and certain kinematic conditions of the neighboring vehicle VV1 in the ENV environment, estimated from the first detected object OBJ1 at such a first acquisition time T1

[0107] Optionally, following step 430, a step (not shown in [Fig. 4]) may include verifying the plausibility of detecting a neighboring vehicle VV1 from the detected object OBJ1 and the parameters (Pn, vn, V) determined from the first detected object OBJ1. A plausibility test may then be optionally implemented, so as to, for example: - verify that the current speed vn reflects a nearby body moving in a plausibly similar way, such as a nearby vehicle VV1 and / or in a direction close to that of the main vehicle VP, - verify that a height of the set of pixels forming the first detected object OBJ1 has a plausible height to belong to a neighboring vehicle VV1 (for example, a first detected object OBJ1 corresponding to the movement of a bird captured in the field of view FOV will potentially have a height not belonging to a range of plausible height values ​​determined to correspond to a vehicle).

[0108] At a step 440, the device 2 can check if the neighboring vehicle VV 1 and / or the first detected object OBJ1 corresponds to a vehicle already previously detected by the device 2, so as to associate the first detected object with an object already tracked. In general, at any given acquisition time, the parameters associated with a detected vehicle, as determined in step 430, allow device 2 to estimate an occupancy space of the ENV environment by the detected vehicle. For example, such an occupancy space of the detected vehicle can be estimated from the vehicle's position (e.g., estimated by a classical classifier or determined in step 430) as well as the dimensions associated with this vehicle (e.g., pre-existing dimensions). Such an occupancy space can, for example, correspond to a set of coordinates according to the world coordinate system (X,Y,Z) in the ENV environment associated with the detected vehicle. As an example, with reference to [Fig.[2], the neighboring vehicle VV2 is already detected by device 2 at the first acquisition time Tl, for example by a conventional classifier detecting and tracking the frontal faces of vehicles. An occupancy space corresponding to a set of positions in the ENV environment is then already stored by device 2 at the stage of the first acquisition time Tl, in association with such a neighboring vehicle VV2. Thus, at the first acquisition time Tl, a position associated with an object detected at the first acquisition time Tl, for example the first position associated with the first object detected OBJ1 at step 410, can then be compared to the occupancy space associated with the neighboring vehicle VV2, so as to verify if the object detected at the first acquisition time Tl corresponds to the movement of the neighboring vehicle VV2.

[0109] Thus, in step 440, a detected object check can for example be implemented by comparing the first position associated with the first detected object OBJ1 and / or the set of parameters associated with the neighboring vehicle VV1 estimated from the first detected object OBJ1 with one or more occupancy spaces (i.e., one or more sets of coordinates) respectively associated with one or more bodies (e.g., vehicles) previously detected and stored by device 2.

[0110] If, at step 440, such a first position associated with the first detected object OBJ 1 and / or the current position of the neighboring vehicle VV1 is contained within an occupancy space stored by device 2, then device 2 determines that the first detected object OBJ1 (and therefore the neighboring vehicle VV1) corresponds to a known object of interest (e.g., a vehicle already tracked) of the device, which was already detected at a time prior to the first acquisition time T1. Device 2 then does not need to create a new virtual object associated with the first detected object OBJ1 and can rely on the data already associated with the known object of interest. Steps 400 to 440 then allow tracking of such an object of interest (for example, the neighboring vehicle VV1) associated with the first detected object OBJ1.The tracking of an object of interest includes in particular the estimation of future parameters associated with such an object of interest (for example, a future position, dimensions). futures, a future speed, etc.) and the update of the object of interest. For this, at step 450, the parameters (P„, ^n) associated with the neighboring vehicle VV1 and the first Acquisition times Tl, as determined from the first detected object OBJ1, can be fed to a Kalman filter (for example, an Extended Kalman Filter or EKF). Such a Kalman filter is specifically configured to estimate future (or predicted) parameters (Pi+i, vi+1, Vi+i) of an object of interest from at least some current parameters (Pn, vn, sn) provided by device 2.

[0111] Once an object of interest has been detected and tracked, the Kalman filter or any other data estimator can estimate future parameters associated with the object of interest. For example, at the end of step 450, feeding the parameters (Pn, v', sn) associated with the neighboring vehicle VV1 to the Kalman filter allows tracking of the neighboring vehicle VV1 based on successive predictions by the Kalman filter regarding the position and speed of the neighboring vehicle VV1 in the ENV environment. In particular, a lateral object distance (corresponding to an evolving lateral gap between the main vehicle VP and the tracked neighboring vehicle VV1) can be predicted.Thus, any object of interest tracked by device 2 can be associated, depending on the moment considered, with a given lateral object distance, as well as with a set of predicted parameters allowing estimation of a position and movement of the tracked object of interest relative to the main vehicle VP, as the object of interest moves in the ENV environment.

[0112] At step 470, based on the estimated future parameters for a tracked vehicle, the device can then update the parameters associated with the tracked vehicle corresponding to the occupancy space. Such a step 470 of updating the parameters associated with a vehicle already detected will be described in detail later in the description.

[0113] If, at step 440, the first position associated with the first detected object OBJ1 and / or the current position of the neighboring vehicle VV1 is not within an occupancy space stored by device 2, then device 2 can proceed to step 460 of creating a virtual object OV1 associated with the first detected object OBJ1 (and therefore with the detected neighboring vehicle VV1). An example of a virtual object OV1 created from the first detected object OBJ1 is illustrated in [Fig. 6]. In particular, the virtual object OV1 can be created from the parameters (Pn, sn) determined in step 430. For example, from the coordinates (X_Wb Y_Wb Z_Wi) determined from the first reference point Wi and using the model of camera 1, a first virtual reference point Wi having coordinates (ywi,zwi) in the image frame (y,z) can be placed on the acquired image, as shown in [Fig.6].In particular, the virtual point Wi can be positioned in the image coordinate system (y,z) under several assumptions: . - The first object detected, OBJ1, corresponds to a rear portion of the neighboring vehicle VV1 (e.g., it includes a rear wheel of the neighboring vehicle VV1), - the virtual reference point Wi is considered to be positioned on a lateral edge of the acquired image. In other words, the lateral portion visible in the acquired image is considered to represent the entire length of the neighboring vehicle VV1 in the state considered in the acquired image, and the first reference point Wi is considered to be positioned on the front face of the neighboring vehicle VV1.

[0114] In another embodiment, the first virtual reference point Wi may be placed elsewhere than on the side border of the image, for example inside the image or even outside the image.

[0115] Thus, with reference to [Fig.6], if the pixel located at the bottom left of the acquired image is considered to be the origin of the image coordinate system (y,z) and the neighboring vehicle VV 1 is detected as being on the right lateral side of the main vehicle VP (as illustrated in Figures 2 and 5 to 12), we have ywi= 0, as illustrated in [Fig.6].

[0116] Furthermore, the zwi coordinate of the first virtual reference point Wi along the z-axis can be determined from the z-coordinate of the first position to determine the lateral distance d[at] in the image frame. In other words, the virtual reference point Wi can have the same height along the z-axis as the detected object OBJ1. In one embodiment, the corresponding height zwi of the virtual reference point Wi along the z-axis can be predefined and stored by the device 2 as corresponding to a predefined ground height. In one embodiment, the height zwi can also differ from the height of the detected object OBJ1.

[0117] Starting from such a virtual reference point Wi positioned on the acquired image and the actual dimensions V associated with the vehicle (corresponding here to the pre-existing dimensions L1, L2, H) associated with the neighboring vehicle VV1, the virtual object OV1 can be created, having coordinates (ywi, zwi) in the two-dimensional (y, z) space. In particular, the virtual object OV1 can substantially include the first detected object OBJ1, since the latter is identified as corresponding, at least partially, to the neighboring vehicle VV1 detected on the acquired image. Such a virtual object OV1 is, for example, shown in Figure 6. The virtual object OV1 created then corresponds to a model, on the acquired image (therefore in the image coordinate system (y, z)), of the occupancy space in the first acquired image of the neighboring vehicle VV1 detected from the first detected object OBJ1.The current position Pn of the neighboring vehicle VV1 can, equivalently, refer to the position of the reference point Wi or to the virtual reference position wb.

[0118] Optionally, the coordinates (ywi,zwi) of the virtual object OV1 thus created can be fed to the Kalman filter, in association with the parameters (Pn, v«, ^), the detected neighbor vehicle VV1 and the first acquisition time TL. Thus, predicted parameters associated with the virtual object OV1 created can be estimated as and when The measure of how the object of interest (typically, the neighboring vehicle VV1) corresponding to the virtual object 0V1 moves within the ENV environment. In particular, an object lateral distance d^oyi can be associated with the virtual object OV1. For example, at the first acquisition time Tl that enabled the creation of the virtual object 0V1 from the first detected object OBJ1, the object lateral distance diatov\ Pcut corresponds to the first lateral distance diatj. The value of the object lateral distance d^oyi can then evolve according to successive predictions of the data estimator estimating the motion of the tracked neighboring vehicle VV1 (and modeled by the virtual object OV1).

[0119] Optionally, such a virtual object OV1 associated with a detected neighboring vehicle VV1 can then be displayed on the communication interface 50 corresponding to a display screen, for example by superimposing the virtual object OV1 onto a display of the image stream from camera 1, as illustrated in Figure 6. Thus, a driver of the main vehicle VP can quickly identify the occupancy space of the detected neighboring vehicle VV1 in their field of view (FOV). In particular, like the first position associated with the detected object OBJ1, the lateral distance [ associated with the detected object OBJ1, or the parameters (Pn, V) associated with the neighboring vehicle VV1, such a virtual object OV1, as shown in [Fig.6], is associated with a given acquisition time - here the first acquisition time Tl - linked to a certain positioning and certain kinematic conditions of the neighboring vehicle VV 1 in the ENV environment, estimated from the detected object OBJ1 at such a first acquisition time Tl. .

[0120] The method of anticipating the follow-up of [Fig. 4] at a second acquisition time T2, subsequent to the first acquisition time T1, is now considered. For example, a first implementation of the aforementioned steps having led to the creation of a virtual object OV1 associated with the first acquisition time T1 and a first detected object OBJ1, a second acquisition time T2 is considered. At the second acquisition time T2, a second object OBJ2 can be detected at step 410. Such a second object OBJ2 can, for example, be detected on a second image acquired at step 400 during the second acquisition time T2, the second image being distinct from the first image on which the first object OBJ1 was detected. The detection of such a second object OBJ2 on a second image is, for example, shown in [Fig. 8]. The first object OBJ1 associated with the first acquisition time T1 is also shown there for comparison.

[0121] As previously described, a second lateral distance djat2 associated with the second detected object OBJ2 can be determined in step 420, a second reference point W2 and parameters (p V„'\ S,*) associated with the second detected object OBJ2 can be determined in step 430.

[0122] In step 440, object verification is implemented on the second detected object OBJ2, so as to identify whether the second detected object OBJ2 can be associated with an existing (already created) virtual object. In particular, in step 440, it can be determined that the second detected object OBJ2 at the second acquisition time T2 can be associated, by its parameters (p^, Vn%) determined in step 430, with the movement of the neighboring vehicle VV1 already previously detected by device 2 (i.e., at the first acquisition time T1), and represented by the virtual object OV1, as illustrated in Figures 6, 7 and 9.

[0123] In step 450, the parameters (p*, Vn*, Sn*) associated with the second detected object OBJ2 can also be fed to the Kalman filter. Thus, the parameters predicted (Pn+i, vn+i, sn+i) by the Kalman filter for the virtual object OVl can notably take into account the parameters (p*, associated with the second object detected OBJ2.

[0124] Step 470, which updates the virtual object OV1 associated with the neighboring vehicle VV1 at the second acquisition time T2, is now detailed. This step 470, which updates the virtual object OV1, allows, at the end of step 470, obtaining a set of parameters (pn+r, vn+i*, sn+i*) that characterize the position and movement of the virtual object OV1 (and therefore of the neighboring vehicle VV1) in the ENV environment at the second acquisition time T2. The virtual object OV1 to be updated is associated with current parameters (Pn, v, sn) that also need to be updated.

[0125] Step 470 of updating the virtual object OV1 takes into account, in particular, the predicted (or future) state and parameters (pn+i, vn+b sn+i) of the virtual object OV1 provided by the data estimator. Such predicted parameters (pn+i, vn+b sn+i) reflect an estimate, for example provided by the Kalman filter of device 2, of the state of the neighboring vehicle VV1 as predicted from the data from the first acquisition time TL. In particular, the predicted state of the virtual object OV1 may also include a predicted lateral distance object diat>Ovi, corresponding to a lateral gap estimated via the Kalman filter between the neighboring vehicle VV1 and the main vehicle VP at the second acquisition time T2.

[0126] In the context of the proposed process, the step 470 of updating the virtual object 407 includes substeps 471, 472 of updating a length of the virtual object OV 1 according to a first predefined criterion and substeps 473, 474 of updating a position of the virtual object OV1 according to a second predefined criterion.

[0127] In a substep 471, the predicted, or future, parameters (pn+b vn+b sn+i) of the virtual object OV1 as estimated by the Kalman filter are compared to the actual parameters (Pn, vn, V) of the virtual object OV1. In particular, the actual position pn associated with the virtual object OV1 and the estimated future position pn+i for the virtual object OV1 are compared. If, in substep 471, a first criterion is satisfied following such a comparison between the current position pn and the future position pn+1, an update 472 of the current dimensions sn, and in particular of a current length sn(X), of the virtual object OV1 is implemented. In particular, such a first criterion is satisfied if the predicted future position pn+i for the virtual object OV1 indicates that the virtual object OV1 has moved backward in the ENV environment relative to its current position pn. For example, the virtual object OV1 can be considered to have a backward position relative to its current position pn if the lateral component of its predicted position pn+i(X) is less than the lateral component of its current position pn(X) in the (X,Y,Z) coordinate system.In such a case, device 2 maintains the state of the virtual object OV1 except for its current dimensions sn by implementing a systematic lengthening of the length of the virtual object OV1 associated with the neighboring vehicle VV1.

[0128] In substep 472, the update of a current length sn of the virtual object OV1 is then implemented. The current dimensions V of the virtual object OV1 associated with the neighboring vehicle VV1 are then updated by modifying a length associated with the neighboring vehicle VV1, such a length being, for example, the X-coordinate of the dimensions V: .v„(X)+5^ec: - 5 is a difference in length (expressed in meters), ' 5 j(X)* is the updated length of the neighboring vehicle VV1 in the world frame, - sn(X) is the current length of the neighboring vehicle VV 1 in the world frame, - n (y\ is the coordinate along the X axis of the position of the predicted reference point Gh-A / W2 via the Kalman filter in the world frame, ■ p (%) is the coordinate along the X axis of the position of the first reference point Wi in the world frame.

[0129] In particular, the length sn(X) corresponds to a longitudinal component (along the X axis) of the dimensions associated with the virtual object OV1.

[0130] In other words, substep step 472 includes an extension of the length of the virtual object OV1 modeling the neighboring vehicle VV1. The difference in position of the reference point W associated with a front face of the neighboring vehicle VV1 between an actual position p (xj and a predicted position p is then reported in length 5.

[0131] In particular, such an extension at step 472 is implemented with: — Wjoù: - viV is the updated reference point associated with the updated virtual object OV*, - M i is the first reference point associated with the virtual object OV1 before the update.

[0132] In other words, the update of the length of the neighboring vehicle VV1 in step 472 is implemented while maintaining the position of a front face of the virtual object OV*. Thus, updating the virtual object OV1 to OV* allows for updating the length of the neighboring vehicle VV1 while preventing the virtual object OV1 from "moving backward" in the acquired image.

[0133] In particular, updating the dimensions sn of the virtual object in substep 472 keeps the current speed vn constant.

[0134] Thus, the updated parameters (P^, v»+i 3 Vh-D at the end of substep 472 satisfy: Pn+} " — Prf vn+1 " = Sn+1 ” Sn

[0135] Such an update of the length ^(X) of the virtual object OV1 associated with the neighboring vehicle VV1 in substep 472 is shown in [Fig.7].

[0136] If, at substep 471, the first criterion is not satisfied, no length update is implemented and step 470 of updating the virtual object OV1 continues with substeps 473, 474.

[0137] In a substep 473, an update of the position pn of the virtual object OV1 is implemented if a second criterion is satisfied. Such a second criterion depends in particular on a second comparison between the second lateral distance diat2 determined from the second detected object OBJ2 and the object lateral distance diatjOvi estimated for the virtual object OV1. In particular, if, in substep 473, the second lateral distance diat2 is strictly less than the object lateral distance diat,Ovi, an update substep 474 of the current position pn of the virtual object OV1 is implemented, the virtual object OV1 being repositioned so that its updated object lateral distance diat.ovi* is minimal (i.e., diat>Ovi* = inin(d|.,l 0\ k diat>2)).Indeed, such an update 474 allows for the correction of a lateral object distance diatj0Vi between the neighboring vehicle VV1 and the main vehicle VP, which was overestimated during the previous iteration, and an update of the position of the virtual object OV1 is required. The update of the current position Pn of the neighboring vehicle VV1 then includes, in particular, at least an update of a lateral component p (yj (i.e., along the Y-axis in the world coordinate system (X,Y,Z)) of the current position Pn of the neighboring vehicle VV1. In one embodiment, the update of the current position pn may also include an update of other components of the current position pn, for example, a height component pn(Z).

[0138] In substep 474, if the second criterion is satisfied, the update of such an updated position includes associating an updated reference point wP with the updated virtual object OV*, such an updated reference point w2* being different from the first reference point wi. The determination of the updated reference point w2* then relies on the second lateral distance diat2- The positioning of such an updated reference point hs” respects, in particular, the same constraints on the image as for the first reference point

[0139] In particular, updating the parameters in substep 474 maintains a constant image stream.

[0140] Thus, the updated parameters (P + *, ^+1* ^i*) at the end of substep 474 can satisfy: pn+C * pn+Ÿ pn+l(Yf* w " pn+l (Y)xw=^+i

[0141] In particular, the expression of the updated parameters (Pvn+i) at the end of substep 474 depends on the definition of the position Pn and the reference points wb w2*. For example, if the point Pn corresponds to the position of a first reference point Wi positioned in the middle of the virtual object OV1, then the updated parameters (P+i*, Sn+O) at the end of substep 474 can satisfy the following relation

[0142] In one embodiment, at substep 474, the size can also be update, for example to maintain a constant space occupied by the object virtual up-to-date OV* in the image:

[0143] Such an update 474 of the position of the virtual object OV1 so as to obtain an updated virtual object OV* is illustrated in [Fig. 9], for which the second is satisfied (the second lateral distance diat7 is less than the lateral object distance d lat.OVl-

[0144] Thus, such an update 474 modifies the position pn, the lateral object distance diatjOvi and thus the velocity vn of the virtual object OV1, while keeping a parameter of the virtual object OV1 constant, for example the collision time, the ITTC, or pn(X) *v.

[0145] If, at substep 473, the object lateral distance diat>Ovi is less than the second lateral distance db,t7, the update 474 of the position of the virtual object OV1 and an update of the virtual object OV1 directly on the basis of the predictions (pn+b vn+b sn+i ) estimated by the Kalman filter can be implemented at a classical update step 475 of the virtual object OV1.

[0146] Thus, the early tracking method described in [Fig. 4] allows early tracking of the neighboring vehicle before a frontal face of the neighboring vehicle VV1 is visible in the Field of view (FOV). In particular, the proposed method allows for a specific update of the length and / or position of the virtual object 0V1, which models the neighboring vehicle VV1, in order to compensate for and reduce overestimated values ​​of the lateral distance separating the neighboring vehicle VV1 and the main vehicle VP, which can lead to dangerous situations (for example, an overestimation of the lateral distance positions the neighboring vehicle VV1 further from the main vehicle than it actually is). Furthermore, updating the length of the virtual object OV1 can be implemented even in the absence of new detection (for example, of a second detected object OBJ2).

[0147] Optionally, the advance tracking method as proposed can be implemented in combination with tracking implemented by a classifier. In other words, the advance tracking method described in [Fig. 4] can be implemented even when a frontal face of the neighboring vehicle VV1 becomes visible, for example at the third acquisition time T3 in [Fig. 10]. Alternatively, the advance tracking method can be implemented as long as at least reliable tracking of the neighboring vehicle VV1 is not yet possible, for example, as long as a frontal face of the neighboring vehicle VV1 is not yet visible (for example, in the scenarios of the first and second acquisition times T1, T2).To achieve this, the parameters fed to the Kalman filter in step 440 can be associated with a chosen variance value so that when parameters estimated by a frontal classifier are fed to the Kalman filter, the parameters estimated via the proposed advance tracking method are no longer significant for the filter in predicting the parameters. The method in [Fig. 4] can therefore be iterated as long as images are acquired by device 2, or as long as the parameters fed to the Kalman filter are used to predict the parameters (for example, as long as data from classifiers, having a lower uncertainty value and therefore a lower variance or covariance, for example, have not yet been fed to the Kalman filter).

Claims

Claims

1. Method for tracking at least one neighboring vehicle (VV1) present in an environment (ENV) of a main vehicle (VP), said neighboring vehicle (VV1) and said main vehicle (VP) being motor vehicles, the method being implemented by a device (2) configured to provide a driving assistance function for the main vehicle (VP), said device (2) being connected to at least one camera (1) on board the main vehicle (VP) and capable of acquiring images of a scene surrounding the main vehicle (VP) according to at least one field of view (FOV) and at acquisition times (T1, T2, T3), the method comprising the following steps: - detecting (410), from at least one first image acquired at a first acquisition time (T1), a first object (OBJ1) having a vertical plane extending laterally relative to the main vehicle (VP), - determining, according to a predefined coordinate system (X, Y, Z),at least one piece of data relating to a first position associated with the first detected object (OBJ1), - creating (460) a virtual object (OV1) associated with a set of parameters (pn,vn,sn) comprising at least one current position (pn) associated with the virtual object (OV1), said current position (pn) being estimated (430) from at least said piece of data relating to the first position according to the predefined coordinate system, - estimating (440), from at least said current position of the virtual object (pn), a future position (pn+1) of the virtual object (OV1), and - updating (470, (pn+l*,vn+l*,sn+l*)) the parameters associated with the virtual object (OV1) from at least said future position (pn+1), said updating (470) of the parameters associated with the virtual object (OV1) comprising, if a first criterion is satisfied (471), an update (472, sn+1*) of a current length (sn(X)) of the virtual object (OV1).,

2. The method of claim 1, wherein updating the current length of the virtual object (OV1) comprises extending the length of the virtual object (OV1) by a value determined from a gap (ô) between the current position (pn) and the future position (pn+1), while maintaining the current position (pn) of the virtual object (OV1) constant.

3. A method according to any preceding claim, in in which the update of the current length of the virtual object (OV1) is implemented if a first comparison between the future position (pn+1) and the current position (pn) indicates a backward position of the virtual object (OV1) in the first acquired image compared to the current position (pn).

4. Method according to any one of the preceding claims, wherein said data relating to the first position comprises a first lateral distance (dlat, 1) associated with the first detected object (OBJ1) for the first acquisition time (Tl), said first lateral distance (dlat, 1) being determined from measurements of the first detected object (OBJ1) in the first acquired image.

5. Method according to any one of the preceding claims, in which the future position (pn+1) is associated with a lateral object distance (dlat,OVl) of the virtual object (OV1).

6. Method according to the preceding claim, further comprising: - obtaining, from at least one second image acquired at a second acquisition time (T2) subsequent to the first acquisition time (T1), data relating to a second position associated with a second detected object (OBJ2), said data relating to the second position comprising a second lateral distance (dlat,2) associated with the second detected object (OBJ2) for the second acquisition time (T2), - if the second detected object (OBJ2) is determined to belong to the virtual object (OV1), comparing, in a second comparison, the second lateral distance (dlat,2) and the object lateral distance (dlat,OV1), wherein the updating (470, (pn+l*,vn+l*,sn+l*)) of the parameters associated with the virtual object (OV1) further depends on a result of the second comparison.

7. Method according to claim 6, in which, if the second lateral distance (dlat,2) is strictly less than the object lateral distance (dlat,OVl), the update (470, (pn+l*,vn+l*,sn+l*)) of the parameters associated with the virtual object (OV1) comprises an update (474, pn+1*) of the current position (pn) of the virtual object (OV1), said update (474, pn+1*) of the current position (pn) including at least one update of a lateral component (pn(Y)) of the current position (pn) of the virtual object (OV1) corresponding to a repositioning of the virtual object (OV1).

8. Method according to any one of the preceding claims, wherein the current position (pn) of the virtual object (OV1) is associated with the position of a first reference point (wl) belonging to the virtual object, said position of a first reference point (wl) belonging to the virtual object (OV1) being determined so that, in the predefined coordinate system corresponding to an image reference (x,y) in the acquired images, the first reference point (wl) is positioned on a vertical end border of the first image.

9. Method according to any one of the preceding claims, in which the parameters (pn,vn,sn) associated with the virtual object (OV1) are updated (470, (pn+l*,vn+l*,sn+l*)) while maintaining a parameter of the virtual object (OV1) constant (Pn*Vn).

10. Device (2) configured to provide a driving assistance function for a main vehicle (VP), said device (2) being connected to at least one camera (1) on board the main vehicle (VP) and capable of acquiring images of a scene surrounding the main vehicle (VP) according to at least one field of view (FOV) and at acquisition times (Tl, T2, T3), in which the device (2) comprises at least one processing circuit (20, 30, 40, 50, 60) configured to implement a method for tracking at least one neighboring vehicle (Wl) present in an environment (ENV) of the main vehicle (VP) according to one of claims 1 to 9.

11. Computer program comprising instructions for implementing the method according to one of claims 1 to 9 when this program is executed by at least one processor (21, 31, 41, 51, 61).

Images