METHOD FOR THE DYNAMIC ESTIMATING OF THE PITCHING MOVEMENT OF A MOTOR VEHICLE USING AN IMAGE CAPTURE SENSOR

Description

[Technical area]

[0001] This patent application relates to the field of driver assistance systems using a video camera-type image acquisition sensor, and more particularly to the optimization of the acquisition of information relating to the environment of a motor vehicle by a dynamic pitch estimation process using an image acquisition sensor that is mounted on the vehicle. [State of the prior art]

[0002] Advanced Driver Assistance Systems (ADAS) are becoming increasingly common in vehicles. They offer a growing number of features designed to improve the comfort and safety of vehicle users. Ultimately, the production of fully autonomous vehicles is a possibility.

[0003] Many of the features offered by such systems rely on real-time analysis of the immediate environment of the vehicle in which they are installed. Whether it's assisting with emergency braking or improving a car's trajectory control, constantly taking into account the characteristics of the vehicle's environment is necessary to ensure the functionality of driver assistance systems.

[0004] In this context, one way to acquire information about a vehicle's immediate surroundings is to use image acquisition and analysis tools. A vehicle can be equipped with one or more video cameras designed to observe its environment at all times, in one or more directions. The images collected by the cameras are then processed by the driver assistance system, which can use them to potentially trigger a driver assistance action, such as an audible and / or visual alert, or even trigger braking or a change of trajectory.

[0005] This type of operation assumes that the images captured by the camera(s) reliably deduce the desired relevant information about the vehicle's environment, with the lowest possible computational cost. This implies, in particular, that the acquired images accurately represent the observed reality and are free from any degradation that could distort the driver assistance system's perception of the environment. For certain functionalities, it also implies that the driver assistance system must be able to correctly locate the various objects present in the three-dimensional space in which the vehicle is moving, based on the images acquired by the camera.

[0006] This is the case, for example, in the classic situation where the driver assistance system uses images acquired by a camera mounted on the windshield of a vehicle, near its rearview mirror. This use presupposes, for certain functionalities, knowing the relationship between the spatial coordinates of a point in space and the corresponding point in the image acquired by the camera, which in turn requires knowing the spatial relationship between the camera and the vehicle. To ensure that these relationships are correctly understood, it is common practice to calibrate the camera, more or less regularly. This calibration aims to determine parameters of the camera, known as intrinsic and extrinsic parameters.

[0007] Intrinsic parameters are linked to the camera's specific technical characteristics, such as its focal length, the position of its optical center, or its distortion model. Calibration allows for the estimation of these parameters and is therefore performed at the factory at the end of the camera's production line.

[0008] The extrinsic parameters, on the other hand, correspond to the positioning and orientation of the camera relative to a world frame of reference, assumed to be parallel to the road, subject to any potential tilt of the vehicle relative to the road on which it is traveling. These parameters include the three translations and three rotations necessary to go from the coordinates of a point, expressed in a frame of reference defined above and called the "world frame of reference," to the coordinates of that point expressed in a frame of reference linked to the camera, called the "camera frame of reference." Calibrating these parameters (six in total) then allows the camera's positioning and orientation relative to the road to be estimated.Furthermore, since these parameters are likely to vary more frequently than intrinsic parameters, for example in the event of a change in vehicle load, the calibration of extrinsic parameters can be carried out relatively regularly in order to continuously adjust their values ​​to the observed situation.

[0009] The concept of "calibration" often refers to a variable that fluctuates very little and determines an absolute angle between the camera and the world frame of reference. It is an average, or a nominal value, that does not take into account camera oscillations.

[0010] On distinguishes between end-of-line vehicle calibration, which uses a production target, and so-called "online" calibration, which relies on the optical flow of the camera.

[0011] Finally, the purpose of camera calibration is to facilitate the use of images acquired by the camera in order to improve the spatial localization of the various objects observed.

[0012] Despite this, no matter how regularly the calibration can be performed, it does not allow tracking of the dynamic pitch and roll that generate significant dynamic changes in camera orientation.

[0013] Camera pitch is defined as an angular movement of the camera around a transverse axis perpendicular to the longitudinal trajectory of the motor vehicle, and roll is defined as an angular movement of the camera around a longitudinal axis parallel to the longitudinal trajectory of the motor vehicle; their value is estimated using the optical flow of the camera.

[0014] The concept of dynamic pitch and roll represents the same physical quantity as calibration, that is, the angle of the camera relative to a reference frame, generally the "world" reference frame.

[0015] However, dynamic pitch and roll do not represent an average or nominal values ​​but an instantaneous value that oscillates around the calibration value.

[0016] Typically, in situations such as going over a speed bump, significant acceleration or deceleration, or a sudden change in gradient, the dynamic, or instantaneous, pitch value can temporarily deviate significantly from its calibrated value. Similarly, when the vehicle travels over a speed bump or pothole "asymmetrically"—that is, when only the wheels on one side of the vehicle cross the speed bump or pothole—or in the case of a sharp turn, the dynamic, or instantaneous, roll value can temporarily deviate significantly from its calibrated value.

[0017] However, accurately determining the pitch and roll angles is crucial for pinpointing the location of objects observed by a vehicle's camera. Therefore, even a temporary miscalculation of these angles can affect the precise determination of the position of objects surrounding the vehicle, potentially limiting or even disabling certain driver assistance system functions that rely on this precise determination.

[0018] To estimate the dynamic pitch, which we will simply refer to as pitch in the following, and the dynamic roll, which we will simply refer to as roll in the following, of a camera mounted in a vehicle in a dynamic way, that is to say in order to appreciate the rapid variations linked to punctual changes in the driving environment, we already know the use of the simultaneous localization and mapping method or SLAM from the English "Simultaneous Localization And Mapping".

[0019] This method consists, firstly, of simultaneously creating a map of a location and locating oneself within it. It can also be used to estimate the camera's pitch and roll, notably by tracking certain prominent objects in the image and their movement from one frame to the next. However, it is highly susceptible to detection noise and is relatively limited in terms of the responsiveness with which the pitch and roll values ​​can be effectively updated, especially when a robust estimation is the objective.

[0020] Another known method involves processing images acquired by a vehicle's onboard camera to identify, in each image, the position of the road markings on which the vehicle is traveling, and thus deducing, in real time, the camera's pitch angle. This method can dynamically estimate pitch but requires driving on a road with parallel and recognizable markings.

[0021] Furthermore, in the case of the two methods discussed so far, the estimation of pitch and roll values ​​is only effective when these values ​​remain sufficiently close to the expected values. In other words, as soon as the amplitude of the pitch variation is very large, as is the case, for example, when crossing a speed bump or during strong acceleration, or as soon as the amplitude of the roll variation is very large, as is the case, for example, when crossing a speed cushion or a pothole, the two methods mentioned above may fail to keep up with the changes in the camera angle.

[0022] To overcome the drawbacks of the two methods mentioned, a process for dynamically estimating the pitch and roll of a motor vehicle is known and is described and represented in document EP3579191B1.

[0023] The process described in document EP3579191B1 includes a first step of calculating an extrinsic tilt angle of a camera estimated by integrating the relative motion of the camera at a current time t, and a second step of estimating an extrinsic tilt angle of the camera with divergence correction.

[0024] The term "angle of inclination" refers to both a pitch angle and a roll angle, that is to say the angle between the camera frame and a frame which is generally the world frame.

[0025] The first calculation step determines the angle of inclination using the following equation: i int ( t ) = i ( t - 1) - θ rel ( t ) with i ( t - 1) the absolute extrinsic angle of inclination of the camera at a previous instant t-1, and rel ( t) the relative angle of the camera at the current time t which expresses the angle of inclination of the sensor between said previous time t-1 and the current time t.

[0026] The second step estimates the camera's tilt angle using the following equation: θ t = θ int t + DAC t ∗ θ calib t − θ int t with i ( t ) the absolute extrinsic angle of inclination of the camera, at the current time t, i int ( t ) the extrinsic tilt angle of the camera estimated by integrating the relative motion of the camera at a current time t during the first calculation step above, at the current time t, DAC ( t ) an anti-divergence coefficient between zero and one, at the current time t, and θ caliber ( t ) the extrinsic tilt angle of calibration at the current time t, which is known at each instant and which corresponds to the nominal value of the tilt angle of the camera in the absence of dynamics of the motor vehicle.

[0027] The anti-divergence coefficient is also known by the acronym DAC, for the English term "Drift Avoiding Coefficient".

[0028] Advantageously, the process described in document EP3579191B1 takes into account the movement of the camera and the multitude of road scenarios without diverging, by using the anti-divergence coefficient.

[0029] However, the method described in document EP3579191B1 is not efficient enough in the scenario of a change in road slope, particularly when the vehicle equipped with the camera and the tracked object are located on a section of road with a distinct slope, for example if the tracked object is located on an uphill slope while the vehicle is located at the bottom of that slope.

[0030] Indeed, the estimation of the tilt angle according to the method described in document EP3579191 B1 is based on an integration of the relative movement of the camera, so that the change in slope is not taken into account.

[0031] Conversely, the method described above, which is based on recognizing ground marking lines, is more effective at estimating the camera's pitch angle when there is a change in slope.

[0032] US Patent 6678631 B2 (DELPHI TECH INC [US]) (2004-01-13) discloses an attitude angle estimator (e.g., roll). The attitude angle estimator uses an attitude angular velocity sensor to detect a vehicle's attitude angular velocity, a first accelerometer to detect the vehicle's vertical acceleration, and a second accelerometer to detect the vehicle's horizontal (e.g., lateral) acceleration. A controller determines an acceleration-based attitude angle based on the detected vertical and horizontal accelerations, and further determines an attitude angle estimate based on the angular attitude rate and the acceleration-based attitude angle.

[0033] Patent CN 111323047 A (BEIJING HUAHANG RADIO MEASUREMENT RES INST) (2020-06-23) discloses a method for dynamically compensating the zero error of a gyroscope based on an accelerometer. The gyroscope's zero error is dynamically compensated through real-time and long-term fusion. Compared to static compensation, dynamic compensation eliminates the processes of acquiring the gyroscope's angular velocity error data, offline analysis, model building, and other steps in a static environment. Instead, the accelerometer is used to dynamically compensate and correct the gyroscope's zero error, ensuring real-time and high-precision gyroscope output. [Exposition of the invention]

[0034] The present invention aims in particular to resolve the drawbacks of the aforementioned prior art by improving the method described in document EP3579191B1 by relying on the pitch estimation of the method based on the recognition of ground marking lines.

[0035] This objective, as well as others which will appear upon reading the description that follows, is achieved with a method for estimating the pitch of a motor vehicle by means of at least one image acquisition sensor which is mounted on said motor vehicle according to claim 1.

[0036] Thus, the method according to the invention makes it possible to improve the estimation of the pitch angle of the vehicle by carrying out an estimation of the pitch angle according to two complementary methods and by refining, where necessary, the estimated pitch angle, according to the result of the comparison of the two methods.

[0037] Depending on other optional features of the process according to the invention, taken alone or in combination: The third comparison step consists of examining at least one second condition which is validated if the quality of the estimation of the first pitch angle i 1 ( t ) calculated during the first step, and if the quality of the estimation of the second pitch angle i 2 ( t) calculated during the second step, are each greater than a second predefined threshold, and the fourth step is executed if the first and second conditions are met, and the fifth step is executed if either the first or second condition is not met. This feature allows the pitch angle refinement to be conditioned on a quality threshold for the first and second pitch angles; the third comparison step consists of examining a third condition which is met if the absolute difference between the first pitch angle i 1 ( t ) and a calibration angle θ caliber ( t ) at the current time t is less than the absolute difference between the second pitch angle i 2 ( t ) and said calibration angle θ caliber ( t ) , or if the calibration angle θ caliber ( t ) is between the first pitch angle i 1 ( t ) and the second pitch angle i 2 ( t ), the calibration angle θ caliber ( t ) being known at each instant and corresponding to the nominal value of the sensor's tilt angle in the absence of vehicle dynamics, and the fourth step is executed if all said conditions are met, and the fifth step is executed if at least one of said conditions is not met. This feature makes it possible to condition the pitch angle refinement on a difference between the calibration angle and the first and second pitch angles; the method includes: a sixth comparison step consisting of examining a fourth condition which is met if the first pitch angle i 1 ( t ) is greater than the second pitch angle i 2 ( t ) and the first refined pitch angle i 1 aff ( t) is less than the second pitch angle i 2 ( t ) , or if the first pitch angle i 1 ( t ) is less than the second angle of pitch i 2 ( t ) and the first refined pitch angle i 1 aff ( t ) is greater than the second pitch angle i 2 ( t ) , a seventh step which is executed if the fourth condition is met and which consists of considering said second pitch angle i 2 ( t ) as calculated during the second step as the exit angle, an eighth step which is executed if the fourth condition is invalidated and which consists of considering said first refined pitch angle i 1 aff ( t) as the exit angle. This feature aims to favor the second pitch angle obtained using the second estimation method under certain conditions; the first method for estimating the first pitch angle i 1 ( t ) of the sensor includes at least: a first phase of calculating a pitch angle of the sensor estimated by integrating the relative motion of the sensor i int ( t ) at a current time t, by the following equation: θ int t = θ 1 t − 1 − θ rel t with: i. i 1 ( t - 1) the first pitch angle i 1 ( t ) of the sensor at a previous instant t-1, and ii. rel ( t ) a relative angle of the sensor at the current time t which expresses the angle of the sensor between the previous time t-1 and the current time t, and a second phase of estimating the first pitch angle i 1 ( t ) of the sensor, by the following equation: θ 1 t = θ int t + DAC t ∗ θ calib t − θ int t with: i. i 1 (t ) the first pitch angle of the sensor, at the current time t, ii. i int ( t ) the pitch angle of the sensor estimated by integrating the relative motion of the sensor during the first calculation phase above, at the current time t, iii. DAC ( t ) the anti-divergence coefficient, which is between zero and one at the current time t, iv. θ calib ( t ) a calibration angle at the current time t, which is known at each instant and which corresponds to the nominal value of the sensor's tilt angle in the absence of vehicle dynamics; the anti-divergence coefficient DAC(t) depends partly on the acceleration of the motor vehicle and partly on the relative motion of the sensor between the current time t and a previous time, the second estimation phase including the calculation of the anti-divergence coefficient DAC(t) by the following equation: DAC t = min min DAC rel t , DAC a → t with: - DAC relay ( t) a relative anti-divergence coefficient which depends on the movement of the sensor and which is obtained by the following equation: DAC rel t = DAC MaxRel ∗ 1 − min ∫ t − NRel t θ rel x θAMR dx , 1 with DAC MaxRel a predetermined maximum value of the relative anti-divergence coefficient DAC rel ( t ) , No a predetermined number corresponding to the chosen value of the number of previous moments used to determine the relative anti-divergence coefficient DAC rel ( t ) , θ rel ( t ) the relative angle of the sensor at the current time t, ΘAMR a predetermined empirical maximum value of the relative angle of the sensor at the current time t rel ( t ) , And - DACa ( t ) an anti-divergence coefficient of acceleration which depends on the acceleration of the motor vehicle (10), and which is obtained by the following equation: DAC a → t = DAC max a → ∗ 1 − min ∫ t − Na t a → x a → AMR dx , 1 with DAC max. a predetermined maximum value of the acceleration anti-divergence coefficient DAC a( t ) , Na the predetermined number corresponding to the chosen value of the number of previous instants used to determine the DAC acceleration anti-divergence coefficient a ( t ) , a ( t ) the acceleration of the motor vehicle at time t, AMR a predetermined empirical maximum value of the acceleration a ( t ) of the motor vehicle, and, during the fourth stage, the anti-divergence coefficient DAC(t) is increased by increasing the predetermined maximum value DAC MaxRel of the relative anti-divergence coefficient DAC rel ( t ) and by increasing the predetermined maximum value DAC max. of the acceleration anti-divergence coefficient DAC a ( t ) ; a penalty anti-divergence coefficient DAC pen ( t ) is added to the anti-divergence coefficient DAC(t)when the measured longitudinal acceleration of the motor vehicle is inconsistent with the evolution of the first pitch angle i 1 ( t ) of the sensor; the penalty anti-divergence coefficient DAC pen ( t ) is obtained by the following equation: DAC pen t = max DAC MaxPen a → t ∗ θ int t , 0 with DAC MaxPen a predetermined maximum value of the penalty anti-divergence coefficient DAC pen ( t ) , a ( t ) the longitudinal acceleration of the motor vehicle and i int ( t ) the pitch angle of the sensor estimated by integrating the relative motion of the sensor at a current instant t; the anti-divergence coefficient DAC ( t ) is obtained by the following equation: DAC ( t ) = minutes ( minutes ( DAC rel ( t ) , DAC a ( t )) + DAC pen ( t ) , 1) with DAC rel ( t) the relative anti-divergence coefficient, which depends on the sensor's movement, DAC a ( t ) the anti-divergence coefficient of acceleration and DAC pen ( t ) the anti-divergence penalty coefficient; the quality of the first pitch angle estimation i 1 ( t ) depends on the quality of the relative angle rel ( t ) and the quality of the calibration angle θ caliber ( t ) ; the second method for estimating the second pitch angle i 2 ( t The sensor's function is to process the images acquired by the sensor to recognize, in each image, the shape of the road markings on which the motor vehicle is traveling, and to deduce the estimate of the second pitch angle. i 2 ( t ) ; the second method for estimating the second pitch angle i 2 ( tThe sensor's function consists of: - analyzing at least one image acquired by the sensor at a given time t to recognize road markings in said image, and selecting a plurality of points that belong to the road markings, - defining a vanishing point at the intersection of said markings, and - comparing the position of said vanishing point with the position of the sensor's optical center to estimate the second pitch angle. i 2 ( t ) of the sensor; - the quality of the estimation of the second pitch angle i 2 ( t ) depends on the number of selected points belonging to the ground marking lines and the length of said marking lines visible in the acquired image; - the first pitch angle i 1 ( t) of the sensor is a pitch angle which corresponds to an angular movement of the sensor around a transverse axis that is generally perpendicular to the longitudinal trajectory of the motor vehicle.

[0038] The present invention also relates to a device for dynamically estimating the pitch of a motor vehicle by means of at least one image acquisition sensor which is mounted on said motor vehicle, according to claim 15. [Description of drawings]

[0039] Other features and advantages of the invention will become apparent from the following description, with reference to the attached figures, which illustrate: [ Fig. 1 ] : a schematic view of a motor vehicle equipped with a suspended cabin and a video camera; ] Fig. 2 ] : a flowchart of the steps in the process for dynamically estimating the pitch of the motor vehicle according to the invention; [ Fig. 3] : a schematic view of the motor vehicle of the figure 1 in which the camera presents a first pitch angle of the video camera according to a first estimation method; [ Fig. 4 ] : an example of how camera orientation changes relative to the pitch angle; [ Fig. 5 ] : a schematic view of an image captured by the camera representing road markings; ] Fig. 6 ] : a schematic view similar to that of the figure 3 in which is represented the first pitch angle of the video camera according to a first estimation method and the second pitch angle of the video camera according to a second estimation method.

[0040] In the documents of this application, the terminology longitudinal, vertical and transverse shall be adopted without limitation, with reference to the L, V, T trihedron indicated in figures 1 And 3 à 6, considering that the motor vehicle extends longitudinally and moves longitudinally forwards.

[0041] Across all these figures, identical or similar elements are identified by identical or similar reference symbols across all figures. [Description of the modes of realization]

[0042] We represented at the figure 1 a motor vehicle 10 of the truck type which carries an image acquisition sensor consisting of a video camera 12.

[0043] The video camera 12 is mounted on the front windshield of a driver's cab 13, for example near the rearview mirror, so as to monitor the immediate surroundings of the motor vehicle 10, as illustrated in the figure 1 through the field of view 14 of the video camera 12.

[0044] The digital images acquired by the video camera 12 are transmitted to a processing unit 16 belonging to a driver assistance system, for example to enable the driver assistance system to detect objects and make an active intervention decision in the driving of the motor vehicle 10.

[0045] Also, the driving assistance system includes a dynamic estimation device 18 of the pitch of the motor vehicle 10 which is adapted to implement a method of dynamic estimation of the pitch of the motor vehicle 10 by means of the video camera 12, according to the invention.

[0046] Pitch refers to an absolute extrinsic angle of inclination of the video camera 12, which corresponds to an angular oscillation of the video camera 12 around a transverse axis A that is generally perpendicular to the longitudinal trajectory of the motor vehicle 10, as can be seen in the figure 1 .

[0047] We represented at the figure 1 A first three-dimensional frame RC {x C , y C , z C} which is called the camera frame and is linked to the video camera 12, and a second three-dimensional frame RM {x M , y M , z M} which is called the world frame. The world frame RM is linked to the front axle 20 of the motor vehicle 10 and is projected onto the road.

[0048] In general, pitch angle is understood to be a rotation angle between axes of the camera frame RC and axes of the world frame RM.

[0049] It should be noted that under certain conditions, particularly in the case of passing over a speed bump, the pitch angle is a rotation angle between axes of the RC camera frame and axes of a gravitational frame (not shown), that is to say a frame which has a vertical axis in reference to Earth's gravity.

[0050] Those skilled in the art will appreciate that the pitch represented here and whose estimation is the subject of the invention is extrinsic pitch, in the sense defined above in relation more generally to the extrinsic parameters of the video camera 12.

[0051] The potential variation of the intrinsic parameters, as defined above, is not addressed in the estimation described below. Indeed, these parameters are calibrated and, in general, are not subject to significant variations over short time scales. They can therefore be considered stable with regard to the dynamic phenomena that the estimation method according to the invention aims to track, which are related to driving events such as driving over a speed bump or a dip in the road, i.e., over a sudden convex or concave change in elevation in the case of pitching.

[0052] With reference to the figure 2According to a first embodiment of the method for dynamically estimating the pitch of the motor vehicle 10 using the video camera 12 according to the invention, the method comprises a first estimation step E1 which consists of estimating a first pitch angle i 1 ( t ) absolute extrinsic of the video camera 12, illustrated in the figure 3 , following a first estimation method.

[0053] The first method for estimating the first pitch angle i 1 ( t ) incorporates an anti-divergence coefficient DAC ( t ) which aims to limit the divergence in the estimation of the first pitch angle i 1 ( t ) .

[0054] More specifically, the first method for estimating the first pitch angle i 1 ( t) of the video camera 12 includes a first phase of calculating an extrinsic pitch angle of the video camera 12 estimated by integrating the relative motion of the camera i int ( t ) at a current time t, by the following equation: θ int t = θ 1 t − 1 − θ rel t with, as we can see it at the figure 4 , i 1 ( t - 1) the first pitch angle i 1 ( t ) estimated absolute extrinsic of video camera 12 at a previous instant t-1, and rel ( t ) a relative tilt angle of the video camera 12 at the current time t which expresses the angle of the video camera 12 between the previous time t-1 and the current time t.

[0055] By previous instant, we mean for example the instant of the previous image acquisition by the video camera 12 in a sequence of successive acquisitions, or the instant of the acquisition of an image separated from a given number of acquisitions, number chosen for example in advance, in said sequence, or even the instant of the acquisition of a previous image acquired a certain given time before instant t.

[0056] Similarly, the expression "relative camera movement" means the movement of the camera between an image captured at time t and another image captured at a previous time.

[0057] An example of the evolution of the orientation of video camera 12 is given in the figure 4 .

[0058] Furthermore, the first method for estimating the first pitch angle i 1 ( t ) of video camera 12 includes a second phase of estimating the first pitch angle i 1 ( t) of video camera 12, by the following equation: θ 1 t = θ int t + DAC t ∗ θ calib t − θ int t with i 1 ( t ) the first pitch angle of the video camera 12 at the current time t, i int ( t ) the pitch angle of video camera 12 estimated by integrating the relative motion of the camera during the first calculation phase above, at the current time t, DAC ( t ) the anti-divergence coefficient, which is between zero and one at the current time t, θ caliber ( t ) an extrinsic calibration angle at the current time t, which is known at each instant and which corresponds to the nominal value of the tilt angle of the video camera 12 in the absence of dynamics of the motor vehicle 10.

[0059] The value of the extrinsic calibration angle θ caliber ( t) as a function of time can be obtained using any known extrinsic calibration method, such as the SLAM method described previously.

[0060] The extrinsic calibration tilt angle θ caliber ( t ) is a reference angle, it is also possible to use the angle of the video camera 12 decided during the design of the vehicle, or the angle estimated during calibration at the end of the vehicle production line.

[0061] The anti-divergence coefficient DAC ( t ) avoids, as its name suggests, the divergence in the estimation of the first pitch angle i 1 ( t ) .

[0062] Indeed, the first estimated pitch angle i 1 ( t ) may diverge from the true value due to the accumulation over time of errors in estimating the angle estimated by integrating the relative motion of the camera i int ( t ).

[0063] In the absence of the anti-divergence coefficient DAC ( t ) , the potential errors in estimating the relative angle of inclination rel ( t The video camera 12 signals would be summed over time, which could lead to errors in estimating the initial estimated pitch angle. i 1 ( t ) and, in extreme cases, significant discrepancies between the first estimated pitch angle i 1 ( t ) and the actual angle of inclination.

[0064] Depending on the value assigned to the anti-divergence coefficient DAC ( t ) , calculating the angle of inclination i ( t ) is more or less restricted by the calibration angle θ caliber ( t ) , the anti-divergence coefficient DAC ( t ) is said to be more or less conservative.

[0065] In order to estimate the first pitch angle i 1 ( t ) of the video camera 12 can take into account the multitude of road scenarios without diverging, the anti-divergence coefficient DAC ( t ) is calculated, during the second phase of process estimation, by the following equation: DAC t = min min DAC rel t , DAC a → t with: DAC rel ( t ) a relative anti-divergence coefficient which depends on the movement of the video camera 12 and which is obtained by the following equation: DAC rel t = DAC MaxRel ∗ 1 − min ∫ t − NRel t θ rel x θAMR dx , 1 with DAC MaxRel a predetermined maximum value of the relative anti-divergence coefficient DAC rel ( t ) , NRel a predetermined number corresponding to the chosen value of the number of previous moments used to determine the relative anti-divergence coefficient DAC rel ( t ) , θ rel ( t ) the relative angle of the video camera 12 at the current time t, ΘAMRa predetermined empirical maximum value of the relative angle of the video camera 12 at the current time t rel ( t ) .

[0066] The anti-divergence coefficient of acceleration DAC a ( t ) is an anti-divergence coefficient which depends on the acceleration of the motor vehicle 10, and which is obtained by the following equation: DAC a → t = DAC max a → ∗ 1 − min ∫ t − Na t a → x a → AMR dx , 1 with DAC max. a predetermined maximum value of the acceleration anti-divergence coefficient DAC a ( t ) , Na the predetermined number corresponding to the chosen value of the number of previous instants used to determine the anti-divergence coefficient of acceleration DAC a ( t ) , a ( t ) the acceleration of the motor vehicle 10 at time t, AMR a predetermined empirical maximum value of the acceleration a ( t ) of the motor vehicle 10.

[0067] Note that the parameters DAC MaxRel , Nrel, θAMR, DAC maxa , Na And AMR can be modified to adjust the behavior of the algorithm that estimates the first pitch angle i 1 ( t ) of the video camera 12.

[0068] Advantageously, the anti-divergence coefficient DAC ( t ) depends both on the movement of the video camera 12 between the current time t and a previous time t-1 with the relative anti-divergence coefficient DAC rel ( t ) , and also the acceleration of the motor vehicle 10 with the anti-divergence coefficient of acceleration DAC a ( t ) .

[0069] Indeed, when the inter-image movement is high, such as when passing over a pothole or a Berlin cushion, the relative anti-divergence coefficient DAC rel ( t ) tends towards zero and the first pitch angle i 1 ( tThe estimated angle of video camera 12 approaches the angle of video camera 12 estimated by integrating the relative motion of the camera i int ( t ).

[0070] Conversely, when the inter-image movement is small, for example during a long emergency braking maneuver or when going around a roundabout, the relative anti-divergence coefficient DAC rel ( t ) tends towards the predetermined maximum value DAC maxrel of the acceleration anti-divergence coefficient DAC a ( t ) and the estimation of the first pitch angle i 1 ( t ) approaches its reference value, which is the calibration angle θ caliber ( t ) according to the example described here.

[0071] The anti-divergence coefficient of acceleration DAC a ( t ) does not depend on the movement of the video camera 12 but on the acceleration a ( t ) of the motor vehicle 10.

[0072] It should be noted that the acceleration a ( t ) of the motor vehicle 10 is a longitudinal acceleration along the longitudinal trajectory of the motor vehicle 10.

[0073] Making the anti-divergence coefficient dependent on acceleration DAC a ( t ) of the acceleration a ( t ) allows us to take into consideration scenarios in which the cabin of the motor vehicle 10, or more generally the support of the video camera 12, is inclined for a non-negligible time by an inertial force.

[0074] According to an alternative implementation of the first embodiment of the method according to the invention, a penalty anti-divergence coefficient DAC pen ( t ) is added to the anti-divergence coefficient DAC ( t ) so that: DAC t = min min DAC rel t , DAC a → t + DAC pen t , 1 with DAC rel ( t ) the relative anti-divergence coefficient which depends on the movement of the video camera 12, DAC a (t ) the anti-divergence coefficient of acceleration and DAC pen ( t ) the penalty anti-divergence coefficient.

[0075] The penalty anti-divergence coefficient DAC pen ( t ) is added to the anti-divergence coefficient DAC ( t ) when the measured longitudinal acceleration of the motor vehicle 10 is inconsistent with the evolution of the first pitch angle i 1 ( t ) .

[0076] En In effect, the penalty anti-divergence coefficient DAC pen ( t ) is obtained by the following equation: DAC pen t = max DAC MaxPen a → t ∗ θ int t , 0 with DAC maxPen a predetermined maximum value of the penalty anti-divergence coefficient DAC pen ( t ) , a ( t ) the longitudinal acceleration of the motor vehicle 10 and i int ( t) the pitch angle of the video camera 12 estimated by integrating the relative motion of the camera described previously.

[0077] Thus, if the longitudinal acceleration a ( t ) is negative, which characterizes braking of the motor vehicle 10, and if the pitch angle estimated by integrating the relative motion of the camera i int ( t ) is positive, which corresponds to a forward pitch, so the product a ( t ) * i int ( t ) is negative, and the measured acceleration of the motor vehicle 10 is considered consistent with the evolution of the first pitch angle i 1 ( t ) , the penalty anti-divergence coefficient DAC pen ( t ) is set to zero and therefore has no effect.

[0078] Indeed, when the motor vehicle 10 brakes, the vehicle is supposed to roll forward.

[0079] Conversely, if the product a ( t ) * i int ( t ) is positive, then the measured acceleration of the motor vehicle 10 is considered inconsistent with the evolution of the first pitch angle i 1 ( t ) , and the penalty anti-divergence coefficient DAC pen ( t ) is taken into account and it will then act in such a way as to bring back the value of the first pitch angle i 1 ( t ) on that of the extrinsic calibration angle θ caliber ( t ) .

[0080] It should be noted that the penalty anti-divergence coefficient DAC pen ( t ) is saturated by the maximum value DAC maxPen to maintain consistent behavior.

[0081] The method according to the invention includes a second estimation step E2, which can be performed before, after, or during the first estimation step E1, and which consists of estimating a second pitch angle. i 2 ( t ) of video camera 12, following a second estimation method.

[0082] The second method for estimating the second pitch angle i 2 ( t The function of the video camera 12 is to process the images acquired by the video camera 12 to recognize, in each image, the shape of the road markings 21 on which the motor vehicle 10 is traveling, typically a left-hand marking line 22 and a marking line right 24 illustrated at the figure 5 , and to deduce, in real time, a pitch angle of the video camera 12.

[0083] We represented at the figure 6 the first pitch angle i 1 ( t) obtained by the first estimation method and the second pitch angle θ 2 ( t ) obtained by the second estimation method, when the motor vehicle 10, illustrated by a wheel, attacks the base of the slope formed by road 21.

[0084] Note that in this configuration, the second pitch angle θ 2 ( t This allows us to estimate the pitch between video camera 12 and the sloping road 21 visible from camera 12, and not the pitch between video camera 12 and the road 21 assumed to be flat, as estimated by the first pitch angle. i 1 ( t ) obtained by the first estimation method.

[0085] Indeed, in reference to the figure 6 the first pitch angle i 1 ( t ) is defined between the ZC axis of camera 12 and a fictitious X' 1M axis which is globally parallel to the portion of road located before the slope, and the second pitch angle i 2 (t ) is defined between the ZC axis of camera 12 and a fictitious axis X' 2M which is globally parallel to the slope visible by camera 12.

[0086] This type of method, based on recognizing road marking lines, is familiar to those skilled in the art. Therefore, this method will only be briefly described below with an example of its implementation.

[0087] Thus, the second method for estimating the second pitch angle i 2 ( t ) of the video camera 12 consists of analyzing at least one image acquired by the video camera 12 at a time t to recognize and extract the marking lines 22, 24 of the road 21 in said image, and select a plurality of points P which belong to the marking lines 22, 24 extracted previously.

[0088] It is assumed that the extraction of the marking lines 22, 24 was carried out upstream of the first step E1 of the process according to the invention.

[0089] As an example, the extraction of the 22, 24 marking lines can be performed using a deep convolutional neural network. These 22, 24 marking lines, thus detected in the image, are then sampled to be considered as points P belonging to the extracted 22, 24 marking lines.

[0090] Once the marking lines 22, 24 are recognized, a vanishing point O1 is defined at the intersection of the marking lines 22, 24, and the position of the vanishing point O1 is compared with the position of the optical center O2 of the video camera 12 to estimate the second pitch angle θ 2 ( t ) of the video camera 12.

[0091] The second pitch angle i 2 ( t) of the video camera 12 is then expressed as the arc-tangent of the difference in vertical position of the vanishing point O1 estimated by intersection of the two extracted marking lines 22, 24 and the vertical position of the optical center O2 of the video camera 12.

[0092] By way of example, the second method for estimating the second pitch angle i 2 ( t ) of the video camera 12 described here, based on the recognition of the marking lines 22, 24, can be replaced by any other method which is suitable for estimating a pitch angle of the video camera 12 and which is complementary to the first method described above.

[0093] According to the first embodiment of the process according to the invention, the process includes a third comparison step E3 illustrated in the figure 2 , consisting of examining a first condition C1 which is validated if the absolute difference between the first pitch angle i 1 ( t ) and the second pitch angle i 2 ( t ) is greater than or equal to a first predefined threshold.

[0094] If the first condition C1 is validated, in other words if the first condition is met, then a fourth step E4 is executed.

[0095] Conversely, if the first condition C1 is invalidated, in other words if the first condition is not met, then a fifth step E5 is executed instead of the fourth step E4.

[0096] The fourth step, E4, consists of refining the first pitch angle. i 1 ( t ) by increasing the anti-divergence coefficient DAC ( t ) to obtain a first refined pitch angle i 1 aff ( t ) considered as the exit angle.

[0097] The term "exit angle" refers to the estimated pitch angle of the video camera 12, which is given by the method according to the invention.

[0098] To this end, during the fourth step E4, the anti-divergence coefficient DAC ( t ) is increased by increasing the predetermined maximum value DAC MaxRel, of the relative anti-divergence coefficient DAC rel ( t ) and by increasing the predetermined maximum value DAC max. of the acceleration anti-divergence coefficient DAC a ( t ) .

[0099] The fifth step, E5, involves considering the first pitch angle. i 1 ( t ) , as calculated during the first step E1, as the exit angle.

[0100] Thus, according to the present invention, the first pitch angle i 1 ( t) estimated by the first method based on the relative pitch of the video camera 12 is refined with a more conservative parameterization, in the case where the first pitch angle i 1 ( t ) is considered too far from the second pitch angle i 2 ( t ) estimated by the second method based on the marking lines 22, 24 on the ground, which proves to be more reliable in the event of a change in slope of road 21 in particular.

[0101] According to a second embodiment of the method according to the invention, the third comparison step E3 consists of examining a second condition C2 which is validated if the quality of the estimation of the first pitch angle i 1 ( t ) calculated during the first estimation step E1, and if the quality of the second pitch angle estimation i 2 ( t ) calculated during the second E2 estimation step, are each greater than a second predefined threshold.

[0102] If the second condition C2 is validated, as well as the first condition C1, in other words if both conditions C1, C2 are met, then the fourth step E4 described previously is executed.

[0103] Conversely, if one of the first condition C1 and the second condition C2 is not met, then the fifth step E5 described previously is executed instead of the fourth step E4.

[0104] According to a preferred embodiment of the invention, the quality of the estimation of the first pitch angle i 1 ( t ) depends on the quality of the relative angle I am a teacher. ( t ) and the quality of the calibration angle θ caliber ( t ) .

[0105] More specifically, the quality of the estimation of the first pitch angle i 1 ( t) includes a first phase of calculating the quality of the video camera angle 12 estimated by integrating the relative motion of the camera i int ( t ) by the following equation: Qualit é θ int t = 0 , 5 ∗ Qualit é θ 1 t − 1 + 0 , 5 ∗ Qualit é θ rel t with Quality 1 ( t - 1) the quality of the first pitch angle i 1 ( t ) of the video camera 12 at a previous time t-1, calculated at a previous cycle, and Quality rel. ( t ) the quality of the relative angle rel ( t ) of the video camera 12 at the current time t.

[0106] The quality of the relative angle rel ( t )) is obtained by the following equation: Qualit é θ rel t = min Qualit é θ rel 1 t ; Qualit é θ rel 2 t with Quality rel. 1( t ) = 0.8 * QualitySampson + 0.2 * Image Quality

[0107] Sampson's quality can be obtained using the following equation: Qualit é DeSampson = DistanceDeSampsonMax − DistanceDeSampsonActuelle DistanceDeSampsonMax − DistanceDeSampsonMin with DistanceFromSampsonMax a fixed value that defines the maximum value of Sampson, DistanceFromSampsonMina fixed value that defines the minimum value of Sampson and CurrentSampsonDistance the Sampson distance calculated based on the selected model to provide the exit angle.

[0108] Image quality can be obtained using the following equation: Qualit é Image = NombreDeSegmentsConcordants NombreDeSegments with NumberOfSegments the number of segments formed between a point identified on a first image at the previous time t-1 and the same point identified on a second image at the following time t, the images being captured by the video camera 12, and Number of Concordant Segments the number of segments that are consistent with each other and are judged to be faithful to the movement of the video camera 12, as opposed to discordant segments which are judged to be inconsistent.

[0109] Also, the term Quality rel. 2( t ) corresponds to the normalized distribution quality to be comparable to the term Quality rel. 1( t ) .

[0110] To calculate the quality of the distribution, the current image is divided into equal parts, for example, twenty equal parts, and then the number of segments is counted in each part. The number of segments required for a uniform distribution is calculated using the following ratio: NombreDeSegments NombreDeParties .

[0111] Next, we calculate the normalized error with respect to a uniform distribution for each of the parts using the following equation: erreur = NombreDeSegments NombreDeParties − NombreDeSegmentsDansChaquePartie NombreDeSegments NombreDeParties

[0112] Distribution quality is the average of the errors for all parts of the image, such as: Qualit é DeR é partition = ∑ 1 − NombreD ′ erreurs NombreDeParties

[0113] Based on the previous calculations, it is possible to calculate the quality of the estimation of the first pitch angle. i 1 ( t ) with the following equation: Qualit é θ 1 t = Qualit é θ int t + DAC t Qualit é θ calib t − Qualit é θ int t with the anti-divergence coefficient DAC ( t ) as defined previously.

[0114] The quality of the estimation of the second pitch angle i 2 ( t ) , which, as a reminder, is an estimate based on the analysis of the 22, 24 marking lines on the ground, depends on the number of selected P points belonging to the 22, 24 marking lines and the length of the marking lines at 22, 24, for example.

[0115] More specifically, the quality of the estimation of the second pitch angle i 2 ( t ) , can be calculated using the following equation: Qualit é θ 2 t = 0 , 5 ∗ Qualit é NombreDePoints + 0 , 5 ∗ Qualit é LongueurLigne with Qualit é NombreDePoints = min 1 NbrPoints NbrMiniPoints with NbrPoints the number of points P used for the approximation of the marking lines 22, 24, for example three points P for the left marking line 22 and four points P for the right marking line according to the example illustrated in the figure 5 , And NbrMiniPoints the minimum number of points P for the approximation of the marking lines 22, 24, which is a predefined number.

[0116] Furthermore, the quality of the length of the marking lines 22, 24 can be determined by the following equation: Qualit é LongueurLigne = QLongueurLigneD + QLongueurLigneG 2 with QLineLengthD the quality of the length of the straight line 24 which is equal to: TailleVerticaleDeLaLigneDroite TailleVerticaleDeLaR é ggionD ′ int é r ê tDeL ′ image And QLength of Line G the quality of the length of the left line 22 which is equal to: TailleVerticaleDeLaLigneDroite TailleVerticaleDeLaR é gionD ′ int é r ê tDeL ′ image .

[0117] The vertical size V1 of the right line 24, the vertical size V2 of the left line 22, and the vertical size V3 of the region of interest in the image are illustrated in the figure 5 .

[0118] Thus, the second implementation of the method according to the invention allows the execution of the fourth step E4 provided that the quality of the estimation of the first pitch angle i 1 ( t ) and that the quality of the estimation of the second pitch angle i 2 ( t) are deemed sufficient. As a reminder, the fourth step E4 aims to refine the output pitch angle by increasing the anti-divergence coefficient DAC ( t ) .

[0119] According to a third embodiment of the method according to the invention, the third comparison step E3 consists of examining a third condition C3 which is validated if the absolute difference between the first pitch angle i 1 ( t ) and the calibration angle θ caliber ( t ) at the current time t is less than the absolute difference between the second pitch angle i 2 ( t ) and the calibration angle θ caliber ( t ) , or if the calibration angle θ caliber ( t ) , is between the first pitch angle i 1 ( t ) and the second pitch angle i 2 ( t ) . The calibration angle θ caliber ( t) is the one described and used previously for the first method of estimating the first pitch angle i 1 ( t ) of the video camera 12.

[0120] If the three conditions C1, C2, C3 are met, then the fourth step E4 described previously is executed.

[0121] Conversely, if one of the three conditions C1, C2, C3 is not met, then the fifth step E5 described previously is executed instead of the fourth step E4.

[0122] Advantageously, the third embodiment of the invention makes it possible to provide an output pitch angle that is within the range of the first pitch angle. i 1 ( t ) and the second pitch angle i 2 ( t ) .

[0123] In other words, the third embodiment of the invention makes it possible to avoid being too conservative with regard to the calibration angle. θ caliber ( t ).

[0124] According to a fourth embodiment of the process according to the invention, the process includes a sixth comparison step E6 which is executed if the fourth step E4 is executed, as can be seen in the flowchart at the figure 2 .

[0125] The sixth step E6 consists of examining a fourth condition C4 which is met if the first pitch angle i 1 ( t ) is greater than the second pitch angle i 2 ( t ) and if the first pitch angle is refined i 1 aff ( t ) is less than the second pitch angle i 2 ( t ) , or if the first pitch angle i 1 ( t ) is less than the second pitch angle i 2 ( t ) and if the first pitch angle is refined i 1 aff ( t ) is greater than the second pitch angle i 2 ( t ).

[0126] If the fourth condition C4 is met, then a seventh step E7 is executed.

[0127] Conversely, if the fourth condition is not met, then an eighth step E8 is executed, instead of the seventh step E7.

[0128] The seventh step, E7, consists of considering said second pitch angle. i 2 ( t ) , as calculated during the second estimation step E2, as the exit angle.

[0129] The eighth step, E8, involves considering the first refined pitch angle. i 1 aff ( t ) as an exit angle.

[0130] Advantageously, the fourth embodiment of the invention makes it possible to avoid being too conservative, or too conservative, with regard to the calibration angle. θ caliber ( t ) .

[0131] Advantageously, the method according to the invention makes it possible to estimate the pitch angle of the vehicle in a stable manner during phases of low pitch movement of the video camera 12.

[0132] Also, the method according to the invention is reactive in the phases of strong pitching movement of the video camera 12, particularly in the phase of changing slope.

[0133] Moreover, the pitch angle estimated by means of the method according to the invention quickly returns to a value relative to the road, and not to the starting point of integration, after a long inclined section of road.

[0134] Naturally, the invention described above is by way of example. It is understood that a person skilled in the art is capable of carrying out different embodiments of the invention without departing from its scope.

Claims

1. A method for estimating the pitch of a motor vehicle (10) by means of at least one image acquisition sensor (12) which is on board said motor vehicle (10) and which is able to provide images of the route (21) on which the motor vehicle (10) is moving, characterised in that it comprises at least: - a first estimation step (E1) consisting of estimating a first pitch angle θ1(t) of the sensor (12) following a first estimation method which integrates an anti-divergence coefficient DAC(t) aimed at limiting the divergence of the estimation of the first pitch angle θ1(t), so that, when the inter-image movement is small, the estimation of the first pitch angle θ1(t) of the sensor (12) approaches a reference value and so that, when the inter-image movement is large, the estimation of the first pitch angle θ1(t) of the sensor (12) approaches the angle of the sensor (12) estimated by integrating the relative movement of the sensor (12), - a second estimation step (E2) which consists of estimating a second pitch angle θ2(t) of the sensor (12), according to a second estimation method for estimating the pitch between the sensor (12) and the route (21) visible to the sensor (12), - a third comparison step (E3) consisting of examining at least one first condition (C1), which is validated if the absolute difference between the first pitch angle θ1(t) and the second pitch angle θ2(t) is greater than or equal to a predefined first threshold, - a fourth step (E4) which is performed if a criterion is fulfilled, the criterion comprising at least the validation of the first condition (C1), and which consists of refining said first pitch angle θ1(t) by increasing the anti-divergence coefficient DAC(t) to obtain a refined first pitch angle θ1aff(t) considered as the output angle, and - a fifth step (E5), which is performed if the first condition (C1) is invalidated, and which consists of considering said first pitch angle θ1(t) as calculated during the first step (E1) as the output angle.

2. The method according to claim 1, characterised in that the third comparison step (E3) consists of examining at least one second condition (C2) which is validated if the quality of the estimation of the first pitch angle θ1(t) calculated in the first step (E1), and if the quality of the estimation of the second pitch angle θ2(t) calculated in the second step (E2), are each greater than a second predefined threshold, and in that the fourth step (E4) is executed if a criterion is met, the criterion further comprising validation of the second condition (C2), and in that the fifth step (E5) is executed if the first condition (C1) or the second condition (C2) are invalidated.

3. The method according to claim 2, characterised in that the third comparison step (E3) consists of examining a third condition (C3) which is validated if the absolute difference between the first pitch angle θ1(t) and a calibration angle calif(t) at the current time t is less than the absolute difference between the second pitch angle θ2(t) and said calibration angle θcalib(t), or if the calibration angle θcalib(t) is between the first pitch angle θ1(t) and the second pitch angle θ1(t), the calibration angle θcalib(t) being known at each instant and corresponding to the nominal value of the inclination angle of the sensor (12) in the absence of dynamics of the motor vehicle (10), and in that the fourth step (E4) is performed if a criterion is met, the criterion further comprising validation of the third condition (C3), and in that the fifth step (E5) is performed, if at least one of said conditions (C1, C2, C3) is invalidated.

4. The method according to any of the preceding claims, characterised in that it comprises: - a sixth comparison step (E6) consisting of examining a fourth condition (C4) which is validated if the first pitch angle θ1(t) is greater than the second pitch angle θ2(t) and the first refined pitch angle θ1aff(t) is less than the second pitch angle θ2(t), or if the first pitch angle θ1(t) is less than the second pitch angle θ2(t) and the first refined pitch angle θ1aff(t) is greater than the second pitch angle θ2(t), - a seventh step (E7) which is performed if the fourth condition (C4) is validated and which consists of considering the said second pitch angle θ2(t) as calculated during the second step (E2) as the output angle, and - an eighth step (E8) which is performed if the fourth condition (C4) is invalidated and which consists of considering the said refined first pitch angle θ1aff(t) as the output angle.

5. The method according to any of the preceding claims, characterised in that the first method of estimating the first pitch angle θ1(t) of the sensor (12) comprises at least: - a first phase of calculating a pitch angle of the sensor (12) estimated by integrating the relative movement of the sensor (12) θint(t) at a t current time, using the following equation: θ int t = θ 1 t − 1 − θ rel t with: i. θ1(t - 1) the first pitch angle θ1(t) of the sensor (12) at a previous time t-1, and ii. θrel(t) a relative angle of the sensor (12) at the current time t, which expresses the angle of the sensor (12) between the previous time t-1 and the current time t, and - a second phase of estimating the first pitch angle θ1(t) of the sensor (12), using the following equation: θ 1 t = θ int t + DAC t * θ calib t − θ int t with: i. θ1(t) the first pitch angle of the sensor (12) at the current time t, ii. θint(t) the pitch angle of the sensor (12) estimated by integrating the relative movement of the sensor (12) during the previous first calculation phase, at the current time t, iii. DAC(t) the anti-divergence coefficient between zero and one, at the current time t, and iv. θcalib(t) a calibration angle at the current time t, which is known at each moment and corresponds to the nominal value of the angle of inclination of the sensor (12) in the absence of vehicle dynamics (10).

6. The method according to any of the preceding claims, characterised in that the anti-divergence coefficient DAC(t) depends in part on the acceleration of the motor vehicle (10) and in part on the relative movement of the sensor (12) between the current time t and a previous time, the second estimation phase comprising the calculation of the anti-divergence coefficient DAC(t) by the following equation: DAC t = min min DAC rel t , DAC a → t with: - DACrel(t) a relative anti-divergence coefficient that depends on the movement of the sensor (12) and is obtained by the following equation: DAC rel t = DAC MaxRel ∗ 1 − min ∫ t − NRel t θ rel x θAMR dx , 1 where DACMaxRel is a predetermined maximum value of the relative anti-divergence coefficient DACrel(t), Nrel is a predetermined number corresponding to the selected value of the number of previous instants used to determine the relative anti-divergence coefficient DACrel(t), θrel(t) the relative angle of the sensor (12) at the current time t, θAMR a predetermined empirical maximum value of the relative angle of the sensor (12) at the current time t θrel(t), and - DACa(t) an acceleration anti-divergence coefficient that depends on the acceleration of the motor vehicle (10), and which is obtained by the following equation: DAC a → t = DAC max a → ∗ 1 − min ∫ t − Na t a → x a → AMR dx , 1 with DACmaxa a predetermined maximum value of the acceleration anti-divergence coefficientDACa(t), Na, the predetermined number corresponding to the selected value of the number of previous instants used to determine the acceleration anti-divergence coefficient DACa(t), a (t) the acceleration of the motor vehicle (10) at instant t, aAMR a predetermined empirical maximum value of the acceleration a (t) of the motor vehicle (10), and in that, during the fourth step (E4), the anti-divergence coefficient DAC(t) is increased by increasing the predetermined maximum value DACMaxRel of the relative anti-divergence coefficient DACrel(t), and by increasing the predetermined maximum value DACmaxa of the acceleration anti-divergence coefficient DACa(t).

7. The method according to any of the preceding claims, characterised in that a penalty anti-divergence coefficient DACpen(t) is added to the anti-divergence coefficient DAC(t) when the measured longitudinal acceleration of the motor vehicle (10) is inconsistent with the evolution of the first pitch angle θ1(t) of the sensor (12).

8. The method according to claim 7, characterised in that the penalty anti-divergence coefficient DACpen(t) is obtained by the following equation: DACpen(t) = max(DACMaxPen(a(t) * θint(t)), 0) where DACMaxPen is a predetermined maximum value of the anti-divergence penalty coefficient DACpen(t), a(t) the longitudinal acceleration of the motor vehicle (10) and θint(t) is the pitch angle of the sensor (12) estimated by integrating the relative movement of the sensor (12) at a current time t.

9. The method according to claims 6, 7, and 8, characterised in that the anti-divergence coefficient DAC(t) is obtained by the following equation: DAC(t) = min(min(DACrel(t),DACa(t)) + DACpen(t),1) with DACrel(t), the relative anti-divergence coefficient that depends on the movement of the sensor (12), DACa (t) the acceleration anti-divergence coefficient, and DACpen(t), the penalty anti-divergence coefficient.

10. The method according to claim 5, characterised in that the quality of the estimation of the first pitch angle θ1(t) depends on the quality of the relative angle θrel(t) and the quality of the calibration angle θcalib(t).

11. The method according to any of the preceding claims, characterised in that the second method for estimating the second pitch angle θ2(t) of the sensor (12) consists of processing the images acquired by the sensor (12) to recognize, in each image, the shape of marking lines (22, 24) on the route (21) on which the motor vehicle (10) is traveling and to deduce from them the estimate of the second pitch angle θ2(t).

12. Method according to claim 11, characterised in that the second method for estimating the second pitch angle θ2(t) of the sensor (12) consists of: - analysing at least one image acquired by the sensor (12) at a time t to recognize marking lines (22, 24) on the surface (21) of the route in said image, and selecting a plurality of points (P) that belong to the marking lines (22, 24) on the surface of the route, - define a vanishing point (O1) at the intersection of said marking lines (22, 24), and - compare the position of said vanishing point (O1) with the position of the optical centre (O2) of the sensor (12) to estimate the second pitch angle θ2(t) of the sensor (12).

13. The method according to claim 12, characterised in that the quality of the estimation of the second pitch angle θ2(t) depends on the number of points (P) selected belonging to the marking lines (22, 24) on the ground and on the length of said marking lines (22, 24) visible in the acquired image.

14. Method according to any of the preceding claims, characterised in that the first pitch angle θ1(t) of the sensor (12) is a pitch angle that corresponds to an angular movement of the sensor (12) about an axis (A) that is generally perpendicular to the longitudinal trajectory of the motor vehicle (10).

15. A dynamic estimation device (18) for estimating the pitch of a motor vehicle (10) by means of at least one image acquisition sensor (12) which is mounted on said motor vehicle (10), designed to implement the method according to any of the preceding claims, said device comprising at least one image acquisition sensor (12) mounted on said vehicle (10) and a processing unit (16) for determining the first pitch angle θ1(t) and the second pitch angle θ2(f) of the sensor (12).