• 专利附录
  • 专利说明
  • 权利要求
  • 类似技术
  • 同族专利
  • 引用文献
  • 法律状态
  • 优先权
    • 专利摘要:
    - Autonomous method and device for determining a global attitude of a motor vehicle. - The method of autonomously determining a so-called global attitude of a motor vehicle (2) during driving along a path on a road (S), said overall attitude (α) illustrating the angle between an axis longitudinal section of the chassis (X2) of the motor vehicle (2) and a longitudinal axis of the road (X1) of the motor vehicle (2) comprises a measurement step implemented using at least two accelerometers arranged on the chassis and configured to measure a longitudinal acceleration and a vertical acceleration of the motor vehicle (2), and a calculating step of calculating the overall attitude (α) of the vehicle (2) using only the measurement pairs measured at the measurement step, said computation step considering that the vertical acceleration and the longitudinal acceleration are linked together by an affine function whose proportionality coefficient depends on a point attitude for a slope (P) of constant route (S).
    公开号:FR3077792A1
    申请号:FR1851247
    申请日:2018-02-14
    公开日:2019-08-16
    发明作者:Jules Dupont Frederic Andre
    申请人:AML Systems SAS；
    IPC主号:
    专利说明:

    The present invention relates in particular to an autonomous method and device for determining at least one trim of a motor vehicle and for correcting its variations with respect to a reference.
    STATE OF THE ART
    In particular to correct the lighting beam of a headlight of a motor vehicle, it may be necessary to know exactly the variations in the attitude of the motor vehicle with respect to a reference. In the context of the present invention, the attitude (called "overall attitude" below) illustrates the angle between a longitudinal axis of reference of the chassis of the motor vehicle and a longitudinal axis of reference of the road on which the wheels of the axle of the motor vehicle. Indeed, depending on the load distribution in the vehicle and its suspensions, the attitude can vary so that the angle of emission of a light beam can be changed. We know that the regulations prohibit an excessive raising of the lighting beam, under certain road conditions, in particular in low beam, in particular to avoid dazzling of an oncoming vehicle. Consequently, a change in the emission angle of a light beam, in particular due to a particular load distribution in the vehicle, can in certain cases cause unauthorized glare. A correction is therefore necessary in such a case.
    To determine the attitude of the vehicle, attitude sensors are known. At least one, and in general at least two attitude sensors are provided on a motor vehicle. Such attitude sensors generally comprise a measurement calculation element linked to two arms which are articulated and which are linked, respectively, to a calculation element secured to the chassis and to a calculation element secured to the axle. Such attitude sensors present in particular a high cost, an assembly complexity and a large bulk.
    Also, to reduce this cost, it is interesting to be able to determine the trim of the motor vehicle using other means.
    Document EP-2 724 889 discloses a device for controlling a lamp in a motor vehicle. This document describes in particular a method for determining an attitude of the motor vehicle. This process uses, to determine the attitude, the affine relationship which links vertical and longitudinal acceleration measures to each other. Such a method therefore uses data directly measured by an acceleration sensor along three orthogonal axes during at least one acceleration phase of the motor vehicle and a deceleration phase of the motor vehicle.
    However, determining an attitude in a precise manner requires the acquisition and / or processing of data taking into account certain realistic parameters of a motor vehicle (suspensions, vibration of the engine, etc.), which is complex and expensive.
    STATEMENT OF THE INVENTION
    The object of the present invention is to remedy this drawback by proposing a method for determining an overall attitude of a motor vehicle during driving on a road, of the motor vehicle along a trajectory, a method which makes it possible to determine and provide a particularly precise attitude value, and this independently.
    To this end, the method for independently determining at least one overall attitude of a motor vehicle during taxiing along a trajectory, said overall attitude illustrating the angle between a longitudinal reference axis of the chassis of the motor vehicle. and a longitudinal reference axis of the road on which the wheels of the axle of the motor vehicle rest is remarkable that it comprises:
    a measurement step implemented using at least two accelerometers arranged on the chassis, the two accelerometers being configured to measure over time, respectively, a so-called longitudinal acceleration and a so-called vertical acceleration of the motor vehicle, the measurement step consisting in measuring a plurality of successive pairs of measurements, each pair of measurements comprising a measured longitudinal acceleration and a measured vertical acceleration;
    a calculation step implemented by a calculation unit and consisting in calculating the overall attitude of the vehicle using exclusively the pairs of measurements measured in the measurement step, said calculation step considering that the vertical acceleration and the longitudinal acceleration are linked to each other by an affine function whose proportionality coefficient depends on a point attitude for a constant road slope, said calculation step comprising a series of successive sub-steps, implemented iteratively and including:
    • a first calculation sub-step consisting in implementing a linear regression, using a plurality of pairs of measurements in order to determine a proportionality coefficient;
    • a second calculation sub-step consisting in estimating a point basis, from said proportionality coefficient determined in the first calculation sub-step;
    • a third calculation sub-step consisting in calculating a quality estimate value of the point basis estimated in the second calculation sub-step; and • a fourth calculation sub-step consisting in determining, at each iteration, a current global attitude, from the set of estimation quality values and punctual bases calculated and estimated in the previous iterations, the global attitude current calculated at the last iteration representing said global attitude.
    Thus, thanks to the invention, we are able to very precisely determine the overall attitude of a motor vehicle during taxiing along a trajectory by calculating the quality of the estimation of the attitude. In addition, the use of information on the quality of the process makes it possible to determine an overall attitude quickly. These advantages make it possible to remedy the aforementioned drawback.
    Advantageously, the measurement step also comprises a step of correcting the values of the vertical acceleration of the pairs of measurements, implemented by a correction module, in order to obtain and provide, in the step of measuring the couples of corrected measures.
    Furthermore, the calculation step also includes a selection step, implemented by a selection module before the first calculation sub-step, consisting in selecting a plurality of successive pairs of measurements from said pairs of measurements measured in step measurement in order to form a data set, said data set being determined by one or more predetermined selection criteria.
    Advantageously, the criterion or criteria for selecting a set of data from a succession of pairs of measurements comprises at least one of the following criteria:
    a number of successive pairs of measurements is greater than a predetermined number of pairs of measurements;
    a succession of pairs of measurements represents a time range less than a predetermined time limit;
    - its values of the longitudinal acceleration of the successive pairs of measurements are increasing or decreasing;
    - an amplitude of the values of the longitudinal acceleration of the successive pairs of measurements is greater than a predetermined variation value;
    - A tolerance value for the temporal regularity of successive measurements of longitudinal accelerations is less than a predetermined tolerance value.
    In addition, preferably, the calculation step comprises a partitioning step, implemented by a partitioning module between the selection step and the first calculation sub-step, consisting in partitioning the data set into a plurality of data subsets and calculating the average of the data of each data subset, the data averages thus obtained being used in the first calculation sub-step as pairs of measurements.
    In a particular embodiment, the fourth calculation sub-step consists in determining a value of said overall attitude by calculating the weighted average of the point bases, said point plates each being weighted by an associated quality of estimation value, calculated in the third calculation sub-step.
    In another embodiment, the fourth calculation sub-step consists in determining a value of said overall attitude by using a Kalman filter.
    Furthermore, the calculation step comprises a fifth calculation sub-step consisting in calculating an overall estimation quality from the estimation quality values calculated in the previous iterations.
    Thus, the quality of the overall estimate provides information on the quality of the measurement of the overall base associated with it. This self-assessment makes it possible, among other things, to choose whether or not to take into account a measurement of the overall attitude as a function of the quality of overall estimation in order to limit the consequences of unreliable measurements in adjusting the headlights, in difficult conditions.
    In a particular embodiment, the fifth calculation sub-step consists in calculating a value of said overall estimate quality defined by the sum of the estimate quality values calculated in the third calculation sub-step.
    In another embodiment, the fifth calculation sub-step consists in calculating a value of said overall estimate quality defined by the average of the estimate quality values calculated in the third calculation sub-step.
    Preferably, the linear regression implemented in the first sub-step of the calculation step is based on a method of least ppercentile of the square of the residuals.
    In addition, the measurement step also comprises a step of filtering the pairs of measurements, implemented by a filter, and consisting, in order to obtain filtered pairs of measurements, in eliminating disturbances on the measurement of the pairs of measurements , generated by:
    - statistical noise from the measurement unit's accelerometer;
    - the vibrations of the motor vehicle engine.
    Advantageously, the measurement step also consists in measuring, using at least one gyrometer, a gyroscopic value, said gyroscopic value measured in the measurement step being used in the calculation step at least to eliminate changes in the trajectory of the motor vehicle on the road, the gyroscopic value of which is greater than a predetermined threshold value.
    In a particular embodiment, said method further comprises:
    an auxiliary calculation step consisting in calculating a correction value for a lighting angle of a headlight of the motor vehicle, using the overall attitude of the motor vehicle calculated in said calculation step; and
    a transmission step consisting in transmitting this correction value to a corrective calculation element capable of correcting the lighting angle of the headlight of the motor vehicle.
    The present invention also relates to a device for the autonomous determination of at least one so-called overall attitude of a motor vehicle during taxiing along a trajectory on a road.
    According to the invention, said device comprises:
    a measurement unit comprising at least one set of accelerometers, said set of accelerometers comprising at least two accelerometers arranged on the chassis of a motor vehicle, the two accelerometers being configured to measure over time, respectively, an acceleration said longitudinal and a so-called vertical acceleration of the motor vehicle, the measurement unit being configured to measure a plurality of successive pairs of measurements, each pair of measurements comprising a measured longitudinal acceleration and a measured vertical acceleration;
    a calculation unit configured to calculate the overall attitude of the vehicle using exclusively the pairs of measurements measured by the measurement unit, said calculation unit considering that the vertical acceleration and the longitudinal acceleration are linked together by an affine function whose proportionality coefficient depends on a point attitude for a constant road slope, the calculation unit comprising:
    • a first calculation element configured to implement a linear regression, using a plurality of pairs of measurements in order to determine a coefficient of proportionality;
    • a second calculation element configured to estimate a point basis, from said proportionality coefficient determined by the first calculation element;
    • a third calculation element configured to calculate a quality estimate value of the point basis estimated by the second calculation element; and • a fourth calculation element configured to calculate a current global attitude, from the set of estimation quality values and the point bases calculated and estimated during a plurality of successive iterations, the current global attitude calculated at the last iteration representing said global base.
    Advantageously, the calculation unit also comprises a selection module configured to select a succession of pairs of measurements from said pairs of measurements measured by the measurement unit in order to form a set of data, said set of data being determined by a or several predetermined selection criteria and a partitioning module configured to partition the data set into a plurality of data subsets and to calculate the average of the data of each data subset, the data averages thus obtained being used by the first calculation element of the calculation unit as pairs of measurements.
    In a particular embodiment, the device further comprises:
    - an auxiliary calculation unit configured to calculate a correction value for a lighting angle of a headlight of the motor vehicle, using the overall attitude of the motor vehicle calculated by said calculation unit; and
    - A transmission link configured to transmit this correction value to at least one corrector calculation element capable of correcting the lighting angle of the headlight of the motor vehicle.
    Furthermore, the calculation unit also comprises a fifth calculation element configured to calculate an overall estimation quality from the estimation quality values calculated during a plurality of successive iterations.
    Preferably, the measurement unit also comprises at least one of the following elements:
    - a third accelerometer configured to measure a lateral acceleration, which is orthogonal to said vertical and longitudinal accelerations;
    - at least one gyrometer.
    Furthermore, the invention also relates to a system for correcting a lighting angle of a headlight of a motor vehicle, said system comprising a device such as that described above, and at least one corrector calculation element capable of correct the lighting angle of the headlight of the motor vehicle.
    In addition, the invention relates to a headlight for a motor vehicle, which is remarkable in that it comprises at least one such lighting angle correction system.
    BRIEF DESCRIPTION OF THE DRAWINGS
    The invention will be better understood, and other objects, details, characteristics and advantages thereof will appear more clearly during the detailed explanatory description which follows, of embodiments of the invention given by way of example. purely illustrative and not limiting, with reference to the accompanying schematic drawings. In these drawings:
    - Figure 1 is the block diagram of a particular embodiment of a device according to the invention;
    - Figure 2 shows a motor vehicle traveling on a slope, to which the invention is applied;
    - Figure 3 is a diagram representing the fluctuations in the value of the attitude around an equilibrium value;
    - Figure 4 is a block diagram of a particular embodiment of a method according to the invention;
    - Figure 5 shows the behavior of the vertical acceleration of the motor vehicle as a function of the longitudinal acceleration; and
    - Figure 6 shows the evolution over time of the point basis, the corresponding quality of estimate and the overall base.
    DETAILED DESCRIPTION
    The device 1 illustrating the invention and shown diagrammatically in FIG. 1 is intended to determine (at least) the so-called overall attitude of a motor vehicle 2 as shown in FIG. 2 and to correct the variations in attitude by compared to a reference.
    In particular, as specified below, this information (overall attitude a) can in particular be used to correct the lighting beam of a headlight 3 of the motor vehicle 2.
    In the context of the present invention, the overall attitude a illustrates the angle between a longitudinal reference axis Χχ of the road on which the wheels of the axle rest (hereinafter "longitudinal road reference axis >>) and a longitudinal reference axis X 2 of the chassis of the motor vehicle 2, as shown in FIG. 2. In general, it is considered that the chassis of the motor vehicle 2 comprising the passenger compartment 4 is connected to the axle provided with the wheels 5 touching route S, via a suspension system. The longitudinal longitudinal axes Xi and X 2 are defined in a vertical plane of symmetry of the motor vehicle 2 with a vertical axis Z 2 , in a horizontal direction when the axle and the chassis are respectively located horizontally. The longitudinal reference axis X 2 of the chassis and the longitudinal road reference axis can be confused when the overall attitude a is zero or angularly separated in the vertical plane of symmetry by an angle corresponding to the overall attitude a , as shown in Figure 2.
    In the example of FIG. 2, the motor vehicle 2 is traveling on a road S along a trajectory in the direction illustrated by an arrow E. The road S is inclined (upwards) by an angle P called a slope , with respect to the horizontal H.
    According to the invention, said device 1 comprises, as shown in FIG. 1:
    a measurement unit 6 comprising a set of accelerometers An. This set comprises at least two accelerometers A1 and A2 linked to the chassis of the motor vehicle 2. The two accelerometers A1 and A2 are configured to measure, respectively, a so-called longitudinal acceleration Ax (which is defined along an axis X corresponding to the longitudinal reference axis X 2 ) and a so-called vertical acceleration Az (which is defined along an axis Z which is orthogonal to the axis X in the vertical plane of symmetry of the motor vehicle 2). In addition, the measurement unit 6 is configured to measure a plurality of successive pairs of measurements, each pair of measurements comprising at least one longitudinal acceleration Ax measured by the accelerometer A1 and a vertical acceleration Az measured by the accelerometer A2; and
    - a calculation unit 7 connected by a link 8 to the measurement unit 6 and configured to calculate the overall attitude a of the motor vehicle 2 using exclusively the measurements carried out by the measurement unit 6.
    The two accelerometers A1 and A2 are arranged substantially orthogonally, rigidly on the chassis of the motor vehicle 2.
    As shown in FIG. 2, a motor vehicle 2 in the running phase on the road S along a trajectory undergoes an acceleration at (Δ v comprising a longitudinal acceleration Ax and a vertical acceleration Az. The longitudinal acceleration Ax and the vertical acceleration Az are measured by the accelerometers A1 and A2 and verify, respectively, the following equations:
    Ax = - cos a. (sin P + Γ) + sin a. (cos P + Z G ) Az = - sin a. (sin P + Γ) + cos a. (cos P + Z G ) where the parameter Z G represents the double time derivative of the distance Z G between the center of gravity G of the motor vehicle 2 and the road S. In the driving phase, the parameter Z G depends on the suspensions of the motor vehicle 2, and the acceleration Γ which it undergoes.
    Cos, sin and tan are the cosine, sine and tangent functions respectively.
    Thus, the longitudinal acceleration Ax and the vertical acceleration Az are linked together by the following expression:
    (cos P + Z G )
    Az = - tan a .Ax -t-cos a
    As shown in Figure 3, the overall trim has fluctuated around a trim value of 0 . These fluctuations δα can be caused by accelerations of the motor vehicle 2, braking of the motor vehicle 2 and / or by a change in the value of the slope P of the road S. The overall attitude has verified the following relationship:
    a = a 0 + δα where δα = μ 3 . (Γ + sinP). The parameter μ β depends, among other things, on the mass of the motor vehicle 2, and represents the effect of the suspensions. The value of the parameter μ 3 is known and can, in particular, be supplied by the car manufacturer. This parameter μ 3 appears in the non-linear relationship between the vertical acceleration Az and the longitudinal acceleration Ax:
    + Z G + 1
    As shown in FIG. 1, in a particular embodiment, the measurement unit 6 comprises a correction module C. This correction module C is configured to correct the values of the vertical acceleration Az so that the plurality of couples of measurements measured by the measurement unit 6 include a corrected vertical acceleration and a longitudinal acceleration Ax. The corrected vertical acceleration and the longitudinal acceleration Ax are then linked together by an affine function via a proportionality coefficient depending on the trim value.
    Az = a Q , as specified below: ~ "o (1 + μ 3 ) 2
    Ax + 1
    In addition, the measurement unit 6 comprises a filtering module F carrying out a filtering of the pairs of measurements provided by the correction module C. This filtering module F is configured to eliminate the disturbances in the pairs of measurements and provide couples filtered measures. The disturbances correspond, for example, to frequency resonance phenomena due to the vibrations of the engine of the motor vehicle 2. The disturbances can also come from the statistical noise of the measurements carried out by the accelerometers A1 and A2. Disturbances can also come from the presence of passengers in the motor vehicle or from the road condition. To this end, the filter module F comprises a medium filter or a Butterworth filter.
    In addition, the measurement unit 6 also comprises at least one gyrometer G capable of measuring a gyroscopic value. In one embodiment, this gyroscopic value corresponds to an angular speed whose value is compared with a threshold value, for example 0.12 rad / sec. When the value of the angular speed is greater than this threshold value, the measurement unit 6 deduces that the trajectory of the motor vehicle 2 on the road S along the vertical and / or longitudinal axes changes rapidly. These changes in trajectory correspond, among other things, to turns in a roundabout, to crossing a speed bump, ... The measurement unit 6 eliminates the pairs of successive measurements, measured simultaneously by the accelerometers A1 and A2, which represent this type of trajectory change. When the value of the angular speed measured by the gyroscopic G is less than or equal to the threshold value, the measurement unit 6 keeps the pairs of measurements measured simultaneously by the accelerometers A1 and A2.
    In one embodiment, the calculation unit 7 comprises a selection module 9 and a partitioning module 10 arranged successively. The selection module 9 is configured to select a plurality of pairs of measurements from the pairs of measurements filtered by the filter F of the measurement unit 6 in order to form a set of data.
    The formation of a data set depends on one or more predetermined selection criteria. Thus, the number of pairs of successive filtered measurements which constitute a set of data must be greater than a number of pairs of predetermined measurements. Without limitation, the number of predetermined measurement pairs is greater than or equal to 30.
    In addition, the data set includes pairs of filtered measurements measured by the measurement unit 6 during a time range less than a predetermined time limit. For example, the time limit is 20 seconds.
    Furthermore, the values of the longitudinal acceleration of the successive pairs of measurements that the measurement unit 6 keeps must be, for example, increasing. This condition implies a positive jerk (derived from acceleration). The values of the longitudinal acceleration of the successive pairs of measurements, kept by the measurement unit 6, can also all be decreasing. In this case, this condition implies a negative jerk. In addition, the amplitude of the values of the longitudinal acceleration Ax of the pairs of successive measurements must be greater than a predetermined variation value. This predetermined variation value is between 0.01g and 0.1g. This selection criterion makes it possible to ensure that the variations in the acceleration values of the pairs of successive measurements are not due to structural problems of the set of accelerometers An.
    Furthermore, a tolerance value of the temporal regularity of the successive measurements of the longitudinal accelerations Ax must be less than a predetermined tolerance value. This tolerance value is between 0.01 seconds and 0.2 seconds. This selection criterion implies that a set of data can be formed by pairs of measurements which have not been successively measured by the measurement unit 6.
    This tolerance makes it possible to adapt the number of data sets from which the calculation unit 7 calculates a global attitude a precisely as a function of its digital resources.
    The partitioning module 10 is configured to partition the data set into a certain number Np of data subsets each comprising several pairs of filtered measurements. The partitioning module 10 is also configured to calculate the average of the corrected vertical accelerations and the average of the longitudinal accelerations Ax of each data subset. The corrected vertical and longitudinal averaged accelerations Ax thus obtained then form the data used by the calculation unit 7 as pairs of measurements.
    In a preferred embodiment, the calculation unit 7 considers that the corrected vertical acceleration ΐζ and the corresponding longitudinal acceleration Ax are linked together by the affine relation:
    Az = a Ax + b where the coefficient represents the slope of a line L, graphical representation of the previous affine relationship (see Figure 5).
    The calculation unit 7 comprises a calculation element 11 configured to implement a simple linear regression calculation using the data provided by the partitioning module 10. The linear regression calculation is based on a method of least p-percentile of the square of the residuals which consists in calculating the coefficients a and h of the line L (see FIG. 5) for which a percentile of p% of the data minimizes the square of the residuals of the data. In one embodiment, this percentile corresponds to a value of p equal to 50%. The method corresponds to the method of least median of the square of the residuals. In another embodiment, this percentile is replaced by 70%. The method corresponds to a method of least 70 percentile of the square of the residuals.
    The calculation unit 7 also includes a calculation element 12 configured to estimate a so-called point basis from the value of the coefficient a and the parameter μ 3 , via the relation a =
    The calculation unit 7 also includes a calculation element 13 configured to calculate an estimate quality value q of the point basis estimated by the calculation element 12. The estimate quality value is proportional to a function increasing the number of subsets Np provided by the partitioning module 10 and a decreasing function of the maximum of the absolute value of the median of the remaining residues Δ.
    In addition, the calculation element 13 of the calculation unit 7 supplies, to a calculation element 14 of the calculation unit 7, estimation quality values q of the point basis as well as the value of the corresponding point base. The calculation element 14 is then configured to calculate the global base a from the weighted average of the point bases obtained during the previous iterations, the point bases being weighted by the value of the estimation qualities q associated with them. .
    The overall attitude a, determined in the manner specified above by the computing unit 7, can be transmitted to a computing element or user system, via a link 15 (FIG. 1).
    In addition, the calculation element 13 of the calculation unit 7 supplies, to a calculation element 21 of the calculation unit 7, estimation quality values Q. The calculation element 21 is configured to calculating an overall estimation quality value from the estimation quality Q values after a defined number of iterations. For example, the overall estimate quality value is the sum of the Q estimate quality values after 20 iterations. The overall estimate quality value can also correspond to the average of the estimate Q quality values after a defined number of operations.
    Furthermore, in a particular embodiment, the device 1 further comprises, as shown in FIG. 1:
    an auxiliary calculation unit 17 configured to calculate a correction value for a lighting angle of a headlight 3 of the motor vehicle 2 (FIG. 2). To do this, the auxiliary calculation unit 17 uses the overall attitude a of the motor vehicle 2 calculated by the calculation unit 7, which is received via a link 16; and
    - A transmission link 18 configured to transmit this correction value to at least one standard corrector calculation element 19, which is capable of correcting the lighting angle of the projector 3 in accordance with this correction value.
    The auxiliary computing unit 17 can be integrated into the computing unit 7.
    In a particular embodiment, the measurement unit 6 further comprises an accelerometer A3 which is configured to measure a lateral acceleration Ay. This lateral acceleration Ay is, by definition, measured along an axis Y (not shown) which is orthogonal to the axes X 2 and Z 2 The device 1 can be part of a correction system 20 shown partially in FIG. 1 and intended for automatically and autonomously correcting a lighting angle of a headlight 3 of a motor vehicle 2.
    This correction system 20 comprises for this purpose, in addition to a device 1, at least one corrector calculation element 19 capable of correcting, in the usual way, the lighting angle of the headlight 3 of the motor vehicle 2 using a correction value received from said device 1 via the transmission link 18.
    Furthermore, the present invention also relates to a projector 3 which, in addition to its usual means (light source, etc.), comprises at least one such system 20 for lighting angle correction.
    In addition, the accelerometers A1, A2 and A3 are microelectromechanical systems of the MEMS type (“Microelectromechanical System” in English). The accelerometers A1, A2 and A3 can provide information at a rate of 1 kHz and the gyrometer G at a rate of 8 kHz.
    Preferably, the measurement unit 6 uses only a sample of the information from the accelerometers A1, A2, and A3 and from the gyrometer G so as not to saturate its digital resources. As an example, sampling by the measurement unit 6 is carried out at a rate between 50 Hz and 200 Hz.
    In addition, in a preferred embodiment, the accelerometers A1, A2 and A3 (or A1 and A2 only) are part of the same accelerometric measurement system An of the three-axis type (or two axes). In addition, the calculation unit 7 is integrated into an electronic card associated with said measurement system.
    The device 1, as described above, implements, automatically, the general steps of the method shown in FIG. 4:
    a measurement step E1 consisting in measuring, using the accelerometers A1 and A2, a plurality of pairs of successive measurements, over time, each pair of measurements consisting of a vertical acceleration Az measured and of a longitudinal acceleration Ax measured; and
    - a calculation step E2 consisting in calculating the overall attitude a, using the pairs of measurements carried out in the measurement step E1. The calculation step E2 considers that the vertical acceleration Az and the longitudinal acceleration Ax are linked together by an affine function whose proportionality coefficient depends on the overall attitude a.
    As explained above, the accelerometers A1 and A2 respectively measure the longitudinal acceleration Ax and the vertical acceleration Az by taking into account realistic phenomena such as the suspensions of the motor vehicle 2. In order to obtain an affine relationship between the values of the vertical acceleration Az and the longitudinal acceleration Ax, the values of the vertical acceleration Az of the pairs of measurements obtained are corrected, during a correction step E11, in particular by knowing the parameter μ 3 which accounts for effects of motor vehicle suspensions 2.
    The measurement pairs comprising the corrected vertical accelerations and the longitudinal accelerations Ax are then filtered by the filter module F during a filtering step E12 in order to form a plurality of pairs of filtered measurements. The filters used during the filtering step E12 are 2 second medium filters or 5 second medium filters. Medium filters can be replaced by Butterworth filters.
    Measurements of a gyroscopic value, for example an angular speed are carried out by the gyroscope G, simultaneously with the measurements carried out by the accelerometers A1 and A2. If the angular speed measured by the gyroscope G is greater than the predetermined threshold value, the pairs of measurements corresponding to the measurements made by the accelerometers A1 and A2 are eliminated from the plurality of pairs of filtered measurements. On the contrary, if the angular speed measured by the gyroscope G is lower than the predetermined threshold value, the pairs of measurements corresponding to the measurements made by the accelerometers A1 and A2 are preserved. During a selection step E21, certain pairs of filtered measurements, from among the pairs of filtered measurements are selected by the selection module 9, according to whether or not they meet at least one selection criterion from the detailed selection criteria. above. The selection criteria are independent of each other.
    The following successive steps are carried out iteratively.
    During a partitioning step E22, the pairs of filtered and selected measurements form a set of data which is partitioned by the partitioning module 10, into a number of data subsets Np. We then perform an average of the values of the pairs of filtered measurements of each subset. This partitioning step E22 makes it possible to reduce the volume of data P1 and P2 (FIG. 5) to be processed during the calculation step E2. The number of subsets of data Np is determined by the total number of pairs of measurements which form the data set as well as by the maximum amplitude of the values of the longitudinal acceleration Ax.
    A simple linear regression calculation step E23 is then carried out on the corrected vertical averaged accelerations zz and the longitudinal averaged accelerations Ax which form the data set obtained after the partitioning step E22. This calculation step E23 is based on a method of minimizing the p-percentile of the square of the residuals of the data P2. For example, this p-percentile can represent the median. It can also be worth 70%. As shown in Figure 5, the method searches for the coefficients a and b of the line L which minimizes the residual square of 70% of the data (black squares P1). Thus, 30% of the data is not considered (white squares P2). The method of least median of the square of the residuals allows the results to be precise by reducing the impact of the so-called outliers. Outliers are, for example, accumulations of corrected vertical acceleration values zz for the same value of the longitudinal acceleration Ax, when the motor vehicle 2 is stopped on road S or when road S presents a strong variation in slope. The calculation step E23 also determines the maximum deviation of the square of the residues Δ 2 .
    A value of the point basis is then deduced from the result of the proportionality coefficient a during a calculation step E24 via the relation a = Λ ~ a ,. The successive point attitude values are (1 + A a ) 2 H represented by circles in Figure 6.
    During a calculation step E25, an estimation quality value q of the point basis, estimated in the previous calculation step E24, is calculated. The quality of estimation q is obtained from the number of data subsets Np determined during the partitioning step E22 as well as from the maximum deviation of the square of the residues Δ 2 , as detailed above.
    During a calculation step E26, the weighted average of the point bases estimated in the preceding steps E24 is calculated, the value of each point plate being weighted by the value of the corresponding quality of estimate q. The result of the weighted average is the overall attitude a (black dots or squares in Figure 6).
    The global attitude obtained at each iteration of the preceding successive steps is transmitted to the auxiliary calculation unit 13, which, during an auxiliary calculation step, uses the value of the global attitude received to calculate a value of correction of the lighting angle of the headlight 3 of the vehicle 2. This correction value is then transmitted, during a step of transmission to a corrective element, capable of correcting the lighting angle of the headlight 3 of the motor vehicle 2.
    Furthermore, during a calculation step E27, a sum of several estimation quality Q values calculated in the calculation step E25 is calculated. The number of Q estimate quality values corresponds to a defined number of iterations. As an example, this defined number of iterations can be equal to 20. The result of the sum of the estimation quality values Q represents an overall estimation quality value.
    权利要求:
    Claims (19)
    [1" id="c-fr-0001]
    1. Method for autonomous determination of at least one so-called overall attitude of a motor vehicle (2) during taxiing along a trajectory on a road (S), said overall attitude (a) illustrating the angle between a longitudinal frame reference axis (X2) of the motor vehicle (2) and a longitudinal reference axis of the road (S) on which the wheels of the axle (X1) of the motor vehicle (2) rest, said method comprising :
    a measurement step (E1) implemented using at least two accelerometers (A1, A2) arranged on the chassis, the two accelerometers (A1, A2) being configured to measure over time, respectively, a so-called longitudinal acceleration and a so-called vertical acceleration of the motor vehicle (2), the measurement step consisting in measuring a plurality of successive pairs of measurements, each pair of measurements comprising a measured longitudinal acceleration and a measured vertical acceleration;
    a calculation step (E2) implemented by a calculation unit (7) and consisting in calculating the overall attitude (a) of the vehicle (2) using exclusively the pairs of measurements measured in the step of measure (E1), said calculation step (E2) considering that the vertical acceleration and the longitudinal acceleration are linked together by an affine function whose coefficient of proportionality depends on a point attitude for a slope (P) of road (S) constant, characterized in that the calculation step (E2) comprises a series of successive substeps, implemented iteratively and comprising:
    • a first calculation sub-step (E23) consisting in implementing a linear regression, using a plurality of pairs of measurements, in order to determine a coefficient of proportionality;
    • a second calculation sub-step (E24) consisting in estimating a point basis, from said proportionality coefficient determined in the first calculation sub-step (E23);
    • a third calculation sub-step (E25) consisting in calculating an estimate quality value (q) of the point basis estimated in the second calculation sub-step (E24); and • a fourth calculation sub-step (E26) consisting in determining, at each iteration, a current global attitude, from the set of estimation quality values (Q) and the point bases calculated and estimated at the iterations previous, the current global base calculated at the last iteration representing said global base (a).
    [2" id="c-fr-0002]
    2. Method according to claim 1, characterized in that the measurement step (E1) also comprises a correction step (E11) of the values of the vertical acceleration of the measurement pairs, implemented by a correction module ( C), in order to obtain and supply, at the measurement step (E2), pairs of corrected measurements.
    [3" id="c-fr-0003]
    3. Method according to any one of claims 1 and 2, characterized in that the calculation step (E2) also comprises a selection step (E21), implemented by a selection module (9) before the first calculation sub-step (E23), consisting in selecting a plurality of successive measurement pairs from said measurement couples measured in the measurement step (E1) in order to form a data set, said data set being determined by a or several predetermined selection criteria.
    [4" id="c-fr-0004]
    4. Method according to claim 3, characterized in that the criteria or criteria for selecting a set of data from a succession of pairs of measurements comprises at least one of the following criteria:
    a number of successive pairs of measurements is greater than a predetermined number of pairs of measurements;
    a succession of pairs of measurements represents a time range less than a predetermined time limit;
    - values of the longitudinal acceleration of the successive pairs of measurements are increasing or decreasing;
    - an amplitude of the values of the longitudinal acceleration of the successive pairs of measurements is greater than a predetermined variation value;
    - A tolerance value for the temporal regularity of successive measurements of longitudinal accelerations is less than a predetermined tolerance value.
    [5" id="c-fr-0005]
    5. Method according to any one of the preceding claims, characterized in that the calculation step (E2) comprises a partitioning step (E22), implemented by a partitioning module (10) between the selection step (E21) and the first calculation sub-step (E23), consisting of partitioning the data set into a plurality of data subsets and calculating the average of the data of each data subset, the data averages as well obtained being used in the first computation substep (E23) as pairs of measurements.
    [6" id="c-fr-0006]
    6. Method according to any one of the preceding claims, characterized in that said fourth calculation sub-step (E26) consists in determining a value of said overall attitude (a) by:
    - a calculation of the weighted average of the point bases, said point plates each being weighted by an associated quality of estimate value (q), calculated in the third calculation sub-step (E25); or
    - use of a Kalman filter.
    [7" id="c-fr-0007]
    7. Method according to any one of the preceding claims, characterized in that the calculation step (E2) comprises a fifth calculation sub-step (E27) consisting in calculating a quality of global estimate from the quality values of estimate (Q) calculated in previous iterations.
    [8" id="c-fr-0008]
    8. Method according to claim 7, characterized in that said fifth calculation sub-step (E27) consists in calculating a value of said overall estimation quality, defined by:
    - the sum of the estimation quality values (Q) calculated in the third calculation sub-step (E24), or
    - the average of the quality of estimate values (Q) calculated in the third calculation sub-step (E24).
    [9" id="c-fr-0009]
    9. Method according to any one of the preceding claims, characterized in that the linear regression, implemented in the first sub-step of the calculation step (E23), is based on a method of least p-percentile of the residual square.
    [10" id="c-fr-0010]
    10. Method according to any one of the preceding claims, characterized in that the measurement step (E1) also comprises a filtering step (E12) of the measurement pairs, implemented by a filter (F), and consisting , in order to obtain filtered pairs of measurements, to eliminate disturbances on the measurement of the pairs of measurements, generated by:
    - the statistical noise from the accelerometer of the measurement unit (6);
    - the vibrations of the motor vehicle engine (2).
    [11" id="c-fr-0011]
    11. Method according to any one of the preceding claims, characterized in that the measuring step consists in also measuring, using at least one gyrometer (G), a gyroscopic value, said gyroscopic value measured at l the measurement step (E1) being used in the calculation step (E2) at least to eliminate changes in the trajectory of the motor vehicle (2) on the road (S) whose gyroscopic value is greater than a predetermined threshold value.
    [12" id="c-fr-0012]
    12. Method according to any one of the preceding claims, characterized in that it further comprises:
    an auxiliary calculation step (E3) consisting in calculating a correction value for a lighting angle of a headlamp (3) of the motor vehicle (2), using the overall attitude (a) of the motor vehicle (2) calculated in said calculation step (E2); and
    - A transmission step (E4) consisting in transmitting this correction value to a corrector calculation element (15) capable of correcting the lighting angle of the headlight (3) of the motor vehicle (2).
    [13" id="c-fr-0013]
    13. Device for autonomous determination of at least one so-called overall attitude of a motor vehicle (2) during taxiing along a trajectory on a road (S), said overall attitude (a) illustrating the angle between a longitudinal chassis reference axis (X2) of the motor vehicle (2) and a longitudinal reference axis of the road on which the wheels of the axle (X1) of the motor vehicle (2) rest, characterized in that it includes:
    - a measurement unit (6) comprising at least one set of accelerometers (An), said set of accelerometers (An) comprising at least two accelerometers (A1, A2) arranged on the chassis of a motor vehicle (2) , the two accelerometers (A1, A2) being configured to measure over time, respectively, a so-called longitudinal acceleration and a so-called vertical acceleration of the motor vehicle (2), the measurement unit (6) being configured to measure a plurality successive pairs of measurements, each pair of measurements comprising a measured longitudinal acceleration and a measured vertical acceleration;
    - a calculation unit (7) configured to calculate the overall attitude (a) of the vehicle (2) using exclusively the pairs of measurements measured by the measurement unit (6), said calculation unit (7) Whereas the vertical acceleration and the longitudinal acceleration are linked together by an affine function whose coefficient of proportionality depends on a point attitude for a constant slope (P) of road (S), characterized in that the unit calculation (7) includes:
    • a first calculation element (11) configured to implement a linear regression, using a plurality of pairs of measurements in order to determine a proportionality coefficient;
    • a second calculation element (12) configured to estimate a point basis, from said proportionality coefficient determined by the first calculation element (11);
    • a third calculation element (13) configured to calculate a quality of estimate value (Q) of the point basis estimated by the second calculation element (12); and • a fourth calculation element (14) configured to calculate a current global attitude, from the set of estimation quality values (q) and the point bases calculated and estimated during a plurality of successive iterations , the current global base calculated at the last iteration representing said global base (a).
    [14" id="c-fr-0014]
    14. Device according to claim 13, characterized in that the calculation unit (7) also comprises:
    - a selection module (9) configured to select a succession of pairs of measurements from said pairs of measurements measured by the measurement unit (6) in order to form a set of data, said set of data being determined by one or more predetermined selection criteria; and
    - a partitioning module (10) configured to partition the data set into a plurality of data subsets and to calculate the average of the data of each data subset, the data averages thus obtained being used by the first calculation element (11) of the calculation unit (7) as pairs of measurements.
    [15" id="c-fr-0015]
    15. Device according to one of claims 13 and 14, characterized in that the calculation unit (7) also comprises a fifth calculation element (21) configured to calculate an overall estimation quality from the quality values estimation (Q) calculated during a plurality of successive iterations.
    [16" id="c-fr-0016]
    16. Device according to any one of claims 13 to 15, characterized in that it further comprises:
    - an auxiliary calculation unit (17) configured to calculate a correction value for a lighting angle of a headlight (3) of the motor vehicle (2), using the overall attitude (a) of the motor vehicle (2) calculated by said calculation unit (7); and
    - A transmission link (18) configured to transmit this correction value to at least one corrector calculation element (19) capable of correcting the lighting angle of the headlight (3) of the motor vehicle (2).
    [17" id="c-fr-0017]
    17. Device according to any one of claims 13 to 16, characterized in that the measurement unit (6) further comprises at least one of the following elements:
    - a third accelerometer (A3) configured to measure a lateral acceleration, which is orthogonal to said vertical and longitudinal accelerations;
    - at least one gyrometer (G).
    [18" id="c-fr-0018]
    18. System for correcting a lighting angle of a motor vehicle headlamp, characterized in that it comprises a device (1) such as that specified under any one of claims 13 to 17, and at least a corrector calculating element (19) capable of correcting the lighting angle of the projector (3).
    [19" id="c-fr-0019]
    19. Headlight for a motor vehicle, characterized in that it comprises at least one system (20) for lighting angle correction, such as that specified in claim 18.
    类似技术:
    公开号 | 公开日 | 专利标题
    FR3077792A1|2019-08-16|AUTONOMOUS METHOD AND DEVICE FOR DETERMINING A PLATE OF A MOTOR VEHICLE.
    FR2701560A1|1994-08-19|Distance measuring device and vehicle speed control device for maintaining the distance between vehicles.
    EP1819984A1|2007-08-22|Hybrid inertial navigation system based on a kinematic model
    FR3057657A1|2018-04-20|VEHICLE NAVIGATION INSTALLATION AND NAVIGATION METHOD APPLIED BY THE INSTALLATION
    FR2994259A1|2014-02-07|METHOD FOR ESTIMATING THE MASS OF A VEHICLE
    EP3571466A1|2019-11-27|Autonomous method and device for determining a trim of a motor vehicle
    EP1804041A1|2007-07-04|Method for real-time estimation of front force and rear force applied by the ground to a vehicle
    WO2020089230A1|2020-05-07|Method for generating control settings for a motor vehicle
    FR3053492A1|2018-01-05|METHOD AND DEVICE FOR DETERMINING AN ESTIMATE OF THE TOTAL MASS OF A MOTOR VEHICLE
    EP3760506B1|2022-02-16|Method for characterising the state of a road
    EP3080565B1|2020-07-01|Device and method for estimating the total mass of a motor vehicle with onboard calibration of suspension displacement sensors
    EP3077781A1|2016-10-12|Method and device for estimating the weight of a motor vehicle
    FR2899844A1|2007-10-19|Motor vehicle e.g. car, suspension controlling method, involves deducing force value applied to actuator from table, longitudinal speed, clearance speed and humidity state, and controlling actuator according to value
    FR3090544A1|2020-06-26|Device and method for controlling the trajectory of a motor vehicle
    FR3094317A1|2020-10-02|Anticipator module, real-time trajectory control device and associated process
    FR3092914A1|2020-08-21|Method for determining the trajectory of a vehicle comprising four steered wheels
    FR2920047A3|2009-02-20|Road slope estimating method for motor vehicle, involves cognizing/estimating real longitudinal acceleration of vehicle, and considering information for measurement as basic input data for calculation of slope
    FR3104106A1|2021-06-11|Learning module, real-time trajectory control device and associated process
    EP3294600A1|2018-03-21|Identification of pitch and yaw inertia of a motor vehicle
    FR3073961A1|2019-05-24|INTEGRATION OF A ROAD PROFILE DETECTION IN ACTIVE VEHICLE SUSPENSION CONTROL
    FR3079026A1|2019-09-20|METHOD FOR CALIBRATING A GYROMETER EQUIPPED WITH A VEHICLE
    FR3111617A1|2021-12-24|Steering control device of a vehicle
    EP1844962A1|2007-10-17|Method of controlling the suspension of an automobile and system including the control method
    FR3086385A1|2020-03-27|METHOD AND SYSTEM FOR CALIBRATION OF SENSORS ON-BOARD IN A MOVING MOTOR VEHICLE
    FR2925003A3|2009-06-19|Motor vehicle derivative determining method, involves selecting motor vehicle derivative calculation model, and calculating derivative of vehicle according to selected calculation model
    同族专利:
    公开号 | 公开日
    KR20190098718A|2019-08-22|
    EP3527453A1|2019-08-21|
    JP2019200196A|2019-11-21|
    FR3077792B1|2020-10-02|
    CN110155074A|2019-08-23|
    EP3527453B1|2021-03-24|
    引用文献:
    公开号 | 申请日 | 公开日 | 申请人 | 专利标题
    EP2447127A2|2010-10-26|2012-05-02|Koito Manufacturing Co., Ltd.|Vehicle lamp controller, vehicle lamp system, and vehicle lamp control method|
    EP2724889A2|2012-10-24|2014-04-30|Koito Manufacturing Co., Ltd.|Control apparatus for vehicle lamp|
    EP2963385A1|2014-07-03|2016-01-06|Memsic, Inc.|Method and apparatus for determining the inclination of a moving vehicle with respect to the road and for performing dynamic headlight leveling|
    GB2536008A|2015-03-03|2016-09-07|Jaguar Land Rover Ltd|Vehicle state estimation apparatus and method|
    CN111252072A|2020-03-05|2020-06-09|上海中科深江电动车辆有限公司|Method for realizing ramp detection processing aiming at pure electric vehicle|
    CN112061106A|2020-09-15|2020-12-11|中国第一汽车股份有限公司|Automatic driving control method, device, vehicle and storage medium|
    法律状态:
    2019-02-28| PLFP| Fee payment|Year of fee payment: 2 |
    2019-08-16| PLSC| Search report ready|Effective date: 20190816 |
    2020-02-28| PLFP| Fee payment|Year of fee payment: 3 |
    2021-02-26| PLFP| Fee payment|Year of fee payment: 4 |
    优先权:
    申请号 | 申请日 | 专利标题
    FR1851247A|FR3077792B1|2018-02-14|2018-02-14|AUTONOMOUS METHOD AND DEVICE FOR DETERMINING A BASE OF A MOTOR VEHICLE.|
    FR1851247|2018-02-14|FR1851247A| FR3077792B1|2018-02-14|2018-02-14|AUTONOMOUS METHOD AND DEVICE FOR DETERMINING A BASE OF A MOTOR VEHICLE.|
    EP19156679.3A| EP3527453B1|2018-02-14|2019-02-12|Autonomous method and device for determining a global inclination of a motor vehicle|
    KR1020190016738A| KR20190098718A|2018-02-14|2019-02-13|Autonomous method and device for determining a global inclination of a motor vehicle|
    CN201910115547.9A| CN110155074A|2018-02-14|2019-02-14|For determining the automated method and device of the integral pitch degree of motor vehicles|
    JP2019024217A| JP2019200196A|2018-02-14|2019-02-14|Autonomous method and device for obtaining global inclination of motor vehicle|
    [返回顶部]