Method and device for determining an estimate of the total mass of a motor vehicle

Abstract

A method determines the total mass of an automotive vehicle on the basis of data of a communication network and parameters of the vehicle, in which an estimation of the total laden mass (mv,est) of the vehicle, of the speed of the vehicle (vest) and of the slope of the road (αest) is determined at an instant (k) by applying the fundamental equation of dynamics and as a function of the values of the total mass of the vehicle, of the speed of the vehicle and of the slope of the road at a previous instant (k−1).

Claims

1. A device-implemented method for determining the total mass of an automotive vehicle in accordance with (i) data from a communication network and (ii) parameters of the vehicle, wherein, in the method, data is retrieved from a database of parameters of the vehicle, the parameters being hosted in a computer of the vehicle, the method comprising: retrieving, from a settings database, setting parameters including (a) execution time of a method for estimating the mass ΔT, (b) a covariance matrix R of measurement noise, (c) a covariance matrix Q of modelling indeterminacies with regard to speed and slope, and (d) a covariance matrix Q.sub.M of modelling indeterminacies with regard to the mass; retrieving data originating from the communication network, the data including (a) a speed of the vehicle v, (b) a downshift ratio of a transmission γ, (c) a level of depression of a brake pedal, (d) a level of use of a retarder, and (e) an estimation of a torque provided by a heat engine T.sub.e and of mechanical friction; determining a motive force F.sub.t(t) generated by the vehicle in accordance with (a) a motor torque value T.sub.e, (b) the downshift ratio of the transmission γ, (c) an efficiency coefficient of the transmission η.sub.gb, and (d) a radius of a wheel r.sub.w; determining an aerodynamic friction force F.sub.a(t) in accordance with the data retrieved from the database of parameters; determining a rolling resistance force F.sub.r(t) in accordance with (a) a mass of the vehicle m.sub.v, (b) a coefficient of rolling resistance of tires c.sub.r, and (c) a slope of a road α; determining a force induced by gravity F.sub.g(t) in accordance with (a) the mass of the vehicle m.sub.v, (b) a gravity constant g, and (c) the slope of the road α; determining an estimation of, at an instant k, (a) a total laden mass m.sub.v,est of the vehicle, (b) a speed of the vehicle v.sub.est, and (c) a slope of the road α.sub.est by applying an equation of dynamics in accordance with values, at a previous instant k−1, of (a) a total mass of the vehicle, a speed of the vehicle, and (c) a slope of the road; and verifying that a journey of the vehicle has terminated, and, when the journey of the vehicle has terminated, (1) computing an average value of the mass Avg(m.sub.v) of the vehicle at an end of the journey, in accordance with the estimation of the mass of the vehicle m.sub.v,est and (2) telematically communicating the computed average value of the mass Avg(m.sub.v) to an external server, wherein in the determining of the estimation of (a) the total laden mass my est of the vehicle, (b) the speed of the vehicle v.sub.est, and (c) the slope of the road α.sub.est, a first Kalman filter F.sub.1 is applied for filtering of measurement of speed of the vehicle and of the estimation of slope of the road, and a second Kalman filter F.sub.2 is applied for estimating the mass of the vehicle, said first filter and said second filter being executed in parallel so that (a) information is pooled between the first filter and the second filter and (b) variables estimated by the first filter are necessary for the estimation of the second filter, and vice versa, and wherein at the same time as the determining of the estimation of (a) the total laden mass my est of the vehicle, (b) the speed of the vehicle v.sub.est, and (c) the slope of the road α.sub.est, an estimation is determined of two covariance matrices including (1) a first covariance matrix P, of dimension 2×2, expressing covariance of an error in speed and slope and (2) a second covariance matrix P.sub.M, of dimension 1×1, expressing covariance of an error in mass, in accordance with (i) the covariance matrix Q of modelling indeterminacies with regard to speed and slope and (ii) the covariance matrix Q.sub.M of modelling indeterminacies with regard to mass, and wherein the two covariance matrices P and P.sub.M are determined in accordance with (a) an equation of dynamics, (b) values at a previous instant k−1 of (i) a total mass of the vehicle, (ii) a speed of the vehicle, and (iii) a slope of the road, and (c) a Riccati equation whose sought-after variable is a covariance matrix of estimation errors.

2. The device-implemented method according to claim 1, further comprising the step of, prior to the retrieving of the data originating from the communication network, initializing at an initial instant (a) the mass of the vehicle m.sub.v, (b) the speed of the vehicle v, (c) the slope of the road α, and (d) the two covariance matrices P and P.sub.M.

3. The device-implemented method according to claim 1, further comprising the step of, prior to the determining of the estimation, verifying at the instant k that conditions have been fulfilled for undertaking a computation of the estimation of the mass.

4. The device-implemented method of determination according to claim 3, wherein the conditions are (a) that the speed of the vehicle v is positive and (b) that both the position of the brake pedal and the level of use of the retarder are zero.

5. The device-implemented method of determination according to claim 1, wherein a value of an estimation error ε(k) is computed by comparing the value of the estimation of the speed v.sub.est with the value of the speed obtained from the communication network.

6. The device-implemented method of determination according to claim 1, wherein at least one proportionality coefficient is computed in accordance with (a) the covariance matrices P and P.sub.M and of (b) the equation of dynamics.

7. The device-implemented method of determination according to claim 6, wherein the estimation of (a) the speed of the vehicle v.sub.est, (b) the slope of the road α.sub.est, and (c) the mass of the vehicle m.sub.v,est is corrected in accordance with (a) a value of an estimation error ε(k) and (b) the at least one proportionality coefficient, wherein a value of the estimation error ε(k) is computed by comparing the value of the estimation of the speed v.sub.est with the value of the speed obtained from the communication network.

8. A device for determining a total mass of an automotive vehicle in accordance with (a) data from a communication network and (b) parameters of the vehicle, wherein the device is configured to receive, as input, (1) data from a database of parameters of the vehicle, the parameters being hosted in a computer of the vehicle and comprising (a) a mass of rolling elements m.sub.r, (b) a coefficient of rolling resistance of tires c.sub.r, (c) a density of air of the vehicle ρ.sub.a, (d) a frontal surface area of the vehicle A.sub.f; (e) a drag coefficient c.sub.x, and (f) a radius of a wheel r.sub.w of the vehicle, (2) setting parameters from a settings database, the setting parameters comprising (a) an execution time of a method for estimating the mass ΔT, (b) a covariance matrix R of measurement noise, (c) a covariance matrix Q of modelling indeterminacies with regard to speed and slope, and (d) a covariance matrix Q.sub.M of modelling indeterminacies with regard to the mass, and (3) data originating from a communication network, the data including (a) a speed of the vehicle v, (b) a downshift ratio of a transmission γ, (c) a level of depression of a brake pedal, (d) a level of use of a retarder, and (e) an estimation of a torque provided by a heat engine T.sub.e and of mechanical friction, wherein the device is further configured to determine (1) a motive force F.sub.t(t) generated by the vehicle, in accordance with (a) a motor torque value T.sub.e, (b) the downshift ratio of the transmission γ, (c) an efficiency coefficient of the transmission η.sub.gb, and (d) the radius of a wheel r.sub.w of the vehicle, (2) an aerodynamic friction force F.sub.a(t), in accordance with the data received from the database of parameters, a rolling resistance force F.sub.r(t), in accordance with (a) the mass of the vehicle m.sub.v, (b) the coefficient of rolling resistance of the tires c.sub.r, and (c) a slope of a road α, and (4) a force induced by gravity F.sub.g(t), in accordance with (a) the mass of the vehicle m.sub.v, (b) a gravity constant g, and (c) the slope of the road α, wherein the device is further configured to make an estimation of, at an instant k, (a) a total laden mass m.sub.v,est of the vehicle, (b) a speed of the vehicle v.sub.est, and (c) a slope of the road α.sub.est by applying an equation of dynamics in accordance with values, at a previous instant k−1, of (a) a total mass of the vehicle, (b) a speed of the vehicle, and (c) a slope of the road, wherein the device is further configured to (1) compute an average value of the mass Avg(m.sub.v) of the vehicle at an end of a journey in accordance with the estimation of the mass of the vehicle m.sub.v,est and (2) telematically communicate the computed average value of the mass Avg(m.sub.v) of the vehicle to an external server, wherein in the determining of the estimation of (a) the total laden mass my est of the vehicle, (b) the speed of the vehicle v.sub.est, and (c) the slope of the road α.sub.est, a first Kalman filter F.sub.1 is applied for filtering of measurement of speed of the vehicle and of the estimation of slope of the road, and a second Kalman filter F.sub.2 is applied for estimating the mass of the vehicle, said first filter and said second filter being executed in parallel so that (a) information is pooled between the first filter and the second filter and (b) variables estimated by the first filter are necessary for the estimation of the second filter, and vice versa, and wherein at the same time as the determining of the estimation of (a) the total laden mass my est of the vehicle, (b) the speed of the vehicle v.sub.est, and (c) the slope of the road α.sub.est, an estimation is determined of two covariance matrices including (1) a first covariance matrix P, of dimension 2×2, expressing covariance of an error in speed and slope and (2) a second covariance matrix P.sub.M, of dimension 1×1, expressing covariance of an error in mass, in accordance with (i) the covariance matrix Q of modelling indeterminacies with regard to speed and slope and (ii) the covariance matrix Q.sub.M of modelling indeterminacies with regard to mass, and wherein the two covariance matrices P and P.sub.M are determined in accordance with (a) an equation of dynamics, (b) values at a previous instant k−1 of (i) a total mass of the vehicle, (ii) a speed of the vehicle, and (iii) a slope of the road, and (c) a Riccati equation whose sought-after variable is a covariance matrix of estimation errors.

9. The device according to claim 8, wherein the device is further configured to correct the estimation, the correction comprising (a) computing an estimation error ε(k) comparing the value of the estimation of the speed v.sub.est with the value of the speed obtained from the communication network and (b) computing a proportionality coefficient in accordance with the covariance matrices P and P.sub.M and the equation of dynamics.

Description

BRIEF DESCRIPTION OF THE FIGURES

(1) Other aims, characteristics and advantages of the invention will become apparent on reading the following description, given solely by way of nonlimiting example and with reference to the appended drawings in which:

(2) FIG. 1 schematically represents the device for determining an estimation of the total mass of a vehicle according to the invention; and

(3) FIG. 2 illustrates the steps of the method for determining an estimation of the total mass of a vehicle according to the invention.

DETAILED DESCRIPTION

(4) A device 10 for determining an estimation of the mass of an automotive vehicle comprises a module for processing the data 20, a computation module 30 for computing the estimation of the mass of the vehicle and of the slope of the road, an end-of-journey detection module 40, a final computation module 50 for finally computing an average value of the mass of the vehicle at the end of the journey, and a communication module 60 for communicating the average value of the mass of the vehicle.

(5) The module for processing the data 20 receives, as input, data retrieved from a database 21 of parameters of the vehicle, which parameters are hosted in a computer (not represented) of the automotive vehicle. The vehicle parameters transmitted to the processing module comprise: the equivalent mass of the rolling elements m.sub.r, expressed in kg. This mass is obtained on the basis of the characteristics of the corresponding components. Thus, the equivalent mass of the wheels is obtained on the basis of the characteristics of the wheels and tires and the equivalent mass of the inertias of the drivetrain (engine, gearbox, transmission shaft . . . ) is obtained on the basis of the characteristics of the heat engine (capacity and number of cylinders). the coefficient of rolling resistance of tires c.sub.r. Since this coefficient is related to the tires, it evolves over time. In order to obtain the best possible accuracy in the estimation of the mass, provision could be made to estimate this coefficient by a specific estimation method. the coefficients of the aerodynamic resistance, such as the density of the air of the vehicle ρ.sub.a, the frontal surface area of the vehicle A.sub.f, the drag coefficient c.sub.x. the radius of the wheels r.sub.w.

(6) The module for processing the data 20 also receives setting parameters from a settings database 22, such as the execution time of the algorithm for estimating the mass ΔT, the covariance matrix of the measurement noise R, the covariance matrix Q of the modelling indeterminacies with regard to the speed and the slope and the covariance matrix Q.sub.M of the modelling indeterminacies with regard to the mass.

(7) The data processing module 20 furthermore receives data originating from a communication network 23 well known to the person skilled in the art under the acronym “CAN: Controller Area Network”. The data transmitted by the CAN network comprise the speed of the vehicle v, the downshift ratio of the gearbox, the level of depression of the brake pedal, the level of use of the retarder and an estimation of the torque provided by the heat engine and of these mechanical frictions.

(8) The downshift ratio of the gearbox, the motor torque, the mechanical friction torque of the heat engine and the radius of the wheels are used to compute the known so-called “control” variable u.

(9) Finally, the module for processing the data 20 can also receive in real time the position of the vehicle in the form of latitude, longitude and altitude as a function of the satellite data received by the global positioning system (not represented).

(10) The module for processing the data 20 comprises an initialization module 24 for initializing, at the instant t=0, the mass of the vehicle m.sub.v, the speed of the vehicle v, the slope of the road α, as well as two covariance matrices P, of dimension 2×2, expressing the covariance of the error in the speed variable and slope variable and P.sub.M, of dimension 1×1, expressing the covariance of the error in the mass variable.

(11) The module for processing the data 20 also comprises a verification module 25 for verifying, at the instant t=k, that all the conditions are fulfilled for undertaking the computation of the estimation of the mass. Stated otherwise, to undertake the computation of the estimation of the mass, the speed of the vehicle v must be strictly positive and the position of the brake pedal and the level of use of the retarder must be zero. As a variant, it would also be possible to envisage conditioning the computation of the estimation of the mass upon the engaging of a ratio of the gearbox.

(12) If one of these conditions is not satisfied, the module for computing the estimation of the mass will undertake the computation of the mass and of the slope as a function of the data at the instant t=k−1.

(13) The computation module 30 for computing the estimation of the mass of the vehicle and of the slope of the road uses an algorithm with a recursive approach based on a mathematical model of the longitudinal dynamics of the vehicle.

(14) The equation of dynamics for the vehicle which links the loadings acting on the vehicle to the mass of the vehicle makes it possible to obtain equation Eq. 1:

(15) $\begin{array}{cc}\left(m_v+m_r\right).Math.\frac{d}{\mathrm{dt}}v\left(t\right)=F_t\left(t\right)-F_{\mathrm{ext}}\left(t\right)& \left(\mathrm{Eq}.1\right)\end{array}$

(16) With: m.sub.v, the mass of the vehicle, expressed in kg; m.sub.r, the equivalent mass of the rolling elements, expressed in kg; v, the speed of the vehicle, expressed in m.Math.s.sup.−2; F.sub.t(t), the traction force applied to the vehicle, expressed in N; and F.sub.ext(t), the exterior forces applied to the vehicle, expressed in N.

(17) The traction force F.sub.t(t) may be written according to the following equation:
F.sub.t(v)=T.sub.e(t).Math.γ(t).Math.η.sub.gb.Math.r.sub.w  (Eq. 2)

(18) With: T.sub.e, the torque of the heat engine, expressed in N.Math.m; γ, the downshift coefficient of the transmission; η.sub.gb, the efficiency coefficient of the transmission; and r.sub.w, the radius of the wheel, expressed in m.

(19) The set of exterior forces F.sub.ext(t) applied to the vehicle groups together the aerodynamic friction force Fa(t), the rolling resistance force F.sub.r(t) and the force induced by gravity F.sub.g(t).

(20) The aerodynamic friction force Fa(t) depends on the density of the air of the vehicle ρ.sub.a, on the frontal surface area of the vehicle A.sub.f, on the drag coefficient c.sub.x and on the relative forward speed of the vehicle with respect to the air (v−v.sub.wind). The aerodynamic friction force Fa(t) may be written according to the following equation:
F.sub.a(v)=½.Math.ρ.sub.a.Math.A.sub.f.Math.c.sub.x.Math.(v−v.sub.wind).sup.2  (Eq. 3)

(21) The speed relatively of the wind v.sub.wind is not known, in order to compute equation Eq. 3, the assumption is made that this speed is negligible, in so far as it does not have any impact on the accuracy of the estimation of the mass of the vehicle.

(22) The rolling resistance force F.sub.r(t) depends on the mass of the vehicle my, on the coefficient of rolling resistance of the tires c.sub.r and on the slope of the road α, as described in the following equation:
F.sub.r(v)=c.sub.r.Math.m.sub.v.Math.cos(α)  (Eq. 4)

(23) The force induced by gravity F.sub.g(t) depends on the mass of the vehicle m.sub.v, on the gravity constant g and on the slope of the road α, as described in the following equation:
F.sub.g(α)=m.sub.v.Math.g.Math.sin(α)  (Eq. 5)

(24) The computation module 30 comprises a first estimation module 31 for estimating the speed v.sub.est of the vehicle, the slope of the road α.sub.est and the mass of the vehicle m.sub.v,est. The first estimation module 31 comprises a first, so-called “Kalman”, filter F.sub.1 allowing the filtering of the measurement of the speed and of the estimation of the slope of the road travelled and a second, so-called “Kalman”, filter F.sub.2 making it possible to estimate the mass of the vehicle. The two filters are executed in parallel so that the information is pooled between the two filters and the variables estimated by the first filter are necessary for the estimation of the second filter, and vice versa.

(25) The estimation of the speed v.sub.est of the vehicle is computed according to the following equation:

(26) $\begin{array}{cc}{v}_{\mathrm{est}}\left(k\right)=v\left(k-1\right)+\frac{\Delta T}{M\left(k-1\right)+m_r}.Math.\left(u\left(k-1\right)-c_r.Math.m_v\left(k-1\right)-g.Math.m_v\left(k-1\right)\sin \left(\alpha \left(k-1\right)\right)-\frac{1}{2}.Math.{\rho }_a.Math.A_f.Math.c_x.Math.v^2\left(k-1\right)\right)& \left(\mathrm{Eq}.6\right)\end{array}$

(27) The estimation of the slope of the road α.sub.est is computed according to the following equation:
sin(α.sub.est(k))=sin(α(k−1))  (Eq. 7)

(28) The estimation of the mass of the vehicle m.sub.v,est is computed according to the following equation:
m.sub.v,est(k)=m.sub.v(k−1)  (Eq. 8)

(29) The first estimation module 31 determines equations 6 to 8 as a function of equations 1 to 5, of the measured variables and of the values of the mass of the vehicle, of the speed of the vehicle and of the slope of the road which were computed at the instant t=k−1.

(30) The computation module 30 also comprises a second estimation module 32 for estimating the two covariance matrices P.sub.est and P.sub.M,is on the basis of the values at the instant t=k−1, of equations 1 to 5 and of a so-called Riccati equation whose sought-after variable is the covariance matrix of the estimation errors:
P.sub.est(k)=A(k).Math.P(k−1).Math.A(k).sup.T+Q  (Eq. 9)
P.sub.M,est(k)=P.sub.M(k−1)+Q.sub.M  (Eq. 10)

(31) With: A, the tangent linearized dynamic matrix (linearization of equations Eq. 6 Eq. 7 with respect to the speed and the slope) Q, the covariance matrix of the modelling indeterminacies with regard to the speed and the slope and the matrix; and Q.sub.M, the covariance matrix of the modelling indeterminacies with regard to the mass.

(32) The computation module 30 furthermore comprises a correction module 33 for correcting the estimation values computed in equations 6 to 8 comprising a module 34 for computing the estimation error ε(k) comparing the value of the estimation of the speed v.sub.est computed by the first estimation module with the value of the speed v.sub.CAN measured by the CAN network:
ε(k)=v.sub.est(k)−v.sub.CAN(k)  (Eq. 11)

(33) The correction module 33 comprises a module 35 for computing the proportionality coefficients L and L.sub.M, termed “gain of the observer”, computed on the basis of the covariance matrices P.sub.est and P.sub.M,is and of the equations of dynamics 1 to 5:
L(k)=(L.sub.1(k),L.sub.2(k)).sup.T=P.sub.est(k).Math.C.sup.T(k).Math.(C(k).Math.P.sub.est(k).Math.C.sup.T(k)+R).sup.−1  (Eq. 12)
L.sub.M(k)=P.sub.M,est(k).Math.C.sub.M.sup.T(k).Math.(C.sub.M(k).Math.P.sub.M,est(k).Math.C.sub.M.sup.T(k)+R).sup.−1  (Eq. 13)

(34) With: L.sub.1, L.sub.2, defined by equation Eq. 12; and C, C.sub.M, the tangent linearized output matrices (linearization of the output equations of the system with respect to the state.fwdarw.C=(0,1) C.sub.M=Cd.sub.f/d.sub.M, where f=Eq. 6 and Eq. 7)

(35) With:
P(k)=P.sub.est(k)−L(k).Math.C(k).Math.P.sub.est(k)  (Eq. 14)
P.sub.M(k)=P.sub.M,est(k)−L.sub.M(k).Math.C.sub.M(k).Math.P.sub.M,est(k)  (Eq. 15)

(36) The corrected value of the speed v of the vehicle is computed according to the following equation:
v(k)=v.sub.est(k)+L.sub.1(k).Math.ε(k)  (Eq. 16)

(37) The corrected value of the slope of the road α is computed according to the following equation:
sin(α(k))=sin(α.sub.est(k))+L.sub.2(k).Math.ε(k)  (Eq. 17)

(38) The corrected value of the mass of the vehicle my is computed according to the following equation:
m.sub.v(k)=m.sub.v,est(k)+L.sub.M(k).Math.ε(k)  (Eq. 18)

(39) The end-of-journey detection module 40 verifies whether the journey has terminated, for example, with the aid of a means of timeout and of comparing the timeout with a threshold value. If the journey has not terminated, the computation module 30 repeats equations 6 to 18.

(40) When the journey has terminated, the final computation module 50 computes an average value of the mass Avg(m.sub.v) of the vehicle at the end of the journey. The communication module 60 thereafter transmits the average value of the mass Avg(m.sub.v) of the vehicle to a server (not represented), for example outside the vehicle, for example by telematic means.

(41) The flowchart represented in FIG. 2 illustrates the method 100 for determining an estimation of the mass of an automotive vehicle.

(42) During a first step 110, the following data are retrieved: the equivalent mass of the rolling elements m.sub.r; the coefficient of rolling resistance of the tires c.sub.r, the coefficients of the aerodynamic resistance, such as the density of the air of the vehicle ρ.sub.a, of the frontal surface area of the vehicle A.sub.f, of the drag coefficient c.sub.x; the radius of the wheels r.sub.w; and the setting parameters such as the execution time of the algorithm for estimating the mass ΔT, the covariance matrix of the measurement noise R, the covariance matrix Q of the modelling indeterminacies with regard to the speed and the slope and the covariance matrix Q.sub.M of the modelling indeterminacies with regard to the mass.

(43) By way of nonlimiting example, it is possible to also retrieve, in real time, the position of the vehicle in the form of latitude, longitude and altitude as a function of the satellite data received by the GPS global positioning system (not represented).

(44) During a second step 120, at the instant t=0, the mass of the vehicle m.sub.v, the speed of the vehicle v, the slope of the road (α), as well as two covariance matrices P, of dimension 2×2, expressing the covariance of the error in the speed variable and slope variable and P.sub.M, of dimension 1×1, expressing the covariance of the error in the mass variable are initialized.

(45) During a third step 130, data originating from the “CAN” communication network are retrieved, such as the speed of the vehicle v, the downshift ratio of the gearbox, the level of depression of the brake pedal, the level of use of the retarder and an estimation of the torque provided by the heat engine and of these mechanical frictions.

(46) During a fourth step 140, it is verified, at the instant t=k, that all the conditions are fulfilled for undertaking the computation of the estimation of the mass. Stated otherwise, to undertake the computation of the estimation of the mass, the speed of the vehicle v must be strictly positive and the position of the brake pedal and the level of use of the retarder must be zero. As a variant, it would also be possible to envisage conditioning the computation of the estimation of the mass upon the engaging of a ratio of the gearbox.

(47) If one of these conditions is not satisfied, the module for computing the estimation of the mass will undertake the computation of the mass and of the slope as a function of the data at the instant t=k−1. Stated otherwise, the values are pegged at the previous instant k−1.

(48) During a fifth step 151, the estimation of the speed v.sub.est of the vehicle, of the slope of the road α.sub.est and of the mass of the vehicle m.sub.v is computed by using an algorithm with a recursive approach based on a mathematical model of the longitudinal dynamics of the vehicle resulting from equations 1 to 5 hereinabove.

(49) At the same time, during a step 152, the estimation of the two matrices P.sub.est and P.sub.M,is is computed according to equations 9 and 10:
P.sub.est(k)=A(k).Math.P(k−1).Math.A(k).sup.T+Q  (Eq. 9)
P.sub.M,est(k)=P.sub.M(k−1)+Q.sub.M  (Eq. 10)

(50) During a sixth step 160, the value of the estimation error ε(k) is computed by comparing the value of the estimation of the speed v.sub.est computed by the first estimation module with the value of the speed v.sub.CAN measured by the CAN network:
ε(k)=v.sub.est(k)−v.sub.CAN(k)  (Eq. 11)

(51) During a seventh step 170, a proportionality coefficient L, termed “gain of the observer”, is computed according to the above equations 12 to 15. The proportionality coefficient L computed on the basis of the covariance matrices P.sub.est and P.sub.M,is and of the equations of dynamics 1 to 5 and the value of the vehicle speed v.sub.est computed according to equation 6, the value of the slope of the road α.sub.est computed according to equation 7 and the value of the vehicle mass of the vehicle m.sub.v,est computed according to equation 8 are corrected.

(52) During an eighth step 180, it is verified whether the journey has terminated, for example, with the aid of a means of timeout and of comparing the timeout with a threshold value. If the journey has not terminated, steps 130 to 170 are repeated.

(53) When the journey has terminated, an average value of the mass Avg(m.sub.v) of the vehicle at the end of the journey is computed, in step 190, and the average value of the mass Avg(m.sub.v) computed is transmitted, in step 200, to a server (not represented), for example outside the vehicle, for example by telematic means.

(54) Thus, the total mass of a vehicle can be estimated in a reliable and fast manner, doing so even under condition of slope of the road travelled.

(55) The vehicle can be an automotive vehicle or a train coach or a trolley or a lifting apparatus.

(56) By mass, by “total mass” of an automotive vehicle, is meant the mass of the vehicle, including its cargo in the case of a semi-trailer or of articulated vehicles and/or, including its passengers in the case of a cargoless vehicle or of a bus.