FR2926316A1 - Motor or resistant type extra effort detecting method for opening e.g. window glass, of motor vehicle, involves comparing dynamic change of characteristic variable with threshold, and emitting signal indicating detection of extra effort - Google Patents

Abstract

The method involves realizing a measurement of quantities associated to an electric actuator (M), where the quantities are supply current (I-M), control voltage (U-M) and a rotation speed (omega-M) of the actuator. A characteristic variable of an extra effort is measured from the quantities, and the variable is stored in a memory for storing consecutive values of the variable. A dynamic change of the variable is calculated from the consecutive values, the change is compared with a threshold. A signal, indicating a detection of the extra effort, is emitted, in case of exceeding the threshold. Independent claims are also included for the following: (1) a method for controlling an electric actuator that is used for opening and closing an opening of a motor vehicle (2) a motor vehicle comprising an opening.

Description

METHOD OF DETECTION AND METHOD OF CONTROLLING AN OPENING ACTUATOR AND VEHICLE USING SUCH A METHOD

The present invention relates, on the one hand, to a method of detecting an additional force exerted on an opening leaf of a motor vehicle. On the other hand, the invention relates to a method of controlling an electric actuator for moving such an opening, this method implementing the method object of the invention. Furthermore, the invention relates to a vehicle of the automotive type implementing such a method. When an opening is actuated by an electric actuator, it is necessary to integrate safety functions with the control of this actuator, so as to limit the force exerted by the opening on a possible obstacle. This obstacle can be on the opening path of the opening, for example a wall, or on its closing path, for example a body part of a user of the door, such as a finger. Such functions thus ensure the physical security of the opening and / or its actuator, as well as the safety of the user.

As an example of motorized opening on a motor vehicle, there may be mentioned in a non-limiting manner electric windows, a retractable roof panel or sliding side doors. A motor vehicle of the prior art, equipped with such openings and electric actuators that move them, generally implements to control each actuator an obstacle detection method that indicates whether to stop or reverse the movement of the opening. The obstacle detection methods of the prior art seldom realize a direct detection of the obstacle, that is to say by means of a sensor integrated in the opening to identify the obstacle, because it is very expensive to cover all areas at risk on the opening. The detection methods of the prior art thus privilege the indirect detection of the obstacle, as illustrated in FIGS. 1 and 2. In the method of FIG. 1, the speed of the opening v (t) is measured during the actuation of the sash identified on a time axis t. The speed measured vmes is compared to a nominal speed profile vnom and the stop stop of the movement of the opening is controlled when the difference Av between the measured speed vmes and the nominal speed vnom exceeds a predetermined threshold. FR-A-2 572 765 describes a similar detection method, which is illustrated in Figure 2. The acceleration a (t) is measured continuously during its movement. The stop stop of the opening is controlled when the acceleration measured ames becomes negative and exceeds a threshold Aa, that is to say when the movement of the opening has a significant deceleration. Alternatively, the indirect obstacle detection methods can measure a magnitude associated with the operation of the electric motor that drives the opening, such as its power supply current, control voltage or rotation speed. The detection of an obstacle is still performed by comparing the magnitude with a nominal profile or by detecting a sudden dynamic variation of this magnitude.

Figures 3 and 4 illustrate such a detection method by measuring a signal S (t) associated with the electric motor operating the opening. In this case, the signal S (t) is derived from an incremental sensor 32, of optical type, detecting singularities 31 arranged on a coding wheel 30. As shown in the timing diagram of FIG. 4, at each rising edge 41 and at each falling edge 42 of the signal S (t), a time delay not to be exceeded is determined before the next edge of the same direction. This delay is defined as the sum of the duration separating the last two rising or falling edges and of an additional duration whose value is precalibrated according to the deceleration of the opening due to a real obstacle. For this, the duration Ato of the cycle separating the last two rising edges constitutes the increment AS of the signal S (t). The detection methods and control methods of the prior art, however, have several disadvantages due to their detection principle. Firstly, the detection thresholds are difficult to calibrate because they are either based on parameters that are not very representative of the actual movement of the opening, such as the supply current or the control voltage of the actuator , or based on parameters, such as the speed of the opening or the nominal profile of this speed, which are adjustable.

In addition, the detection methods of the prior art do not take into account all measurable quantities and representative of the movement of the opening. The methods of the prior art therefore use only the detection of a single quantity. However, false detections can be induced by the manufacturing disparities of a series of openings, climatic conditions, hygrometry, temperature or wind, and by the inclination of the vehicle in a slope or on a slope. Thus, it is difficult to find a compromise between the need to minimize the consequences of a collision between the opening and the obstacle, such as a pinch or a shock, and the need to avoid false detections, for which there is no obstacle. Detection based on a single size is more prone to false detections. Furthermore, the detection methods of the prior art offer no assistance function to the opening or closing of the opening for the case where the user wishes to accelerate these movements. Such a function is particularly suitable for sliding side doors. The present invention aims in particular to remedy these drawbacks, by proposing a method of detecting an effective electric actuator, not subject to false detections, reliable, that is to say not requiring recalibration, and offering assistance to the user of the opening.

For this purpose, the object of the invention is a method of detecting an additional force or load, which is exerted on an opening leaf of a motor vehicle and which is superimposed on the actuating force of said opening exerted by an actuator. characterized in that it comprises the iterative steps of: - a) making measurements of several quantities associated with the actuator; b) estimating, from the measurements carried out in step a), at least one characteristic variable of said additional effort; - c) registering said variable in a memory able to record several consecutive values of said variable; - d) calculate, from the consecutive values recorded in said memory, the dynamic variation of the variable; e) comparing said variation with at least one threshold; and 30 - f) if said threshold is exceeded, transmit a signal indicating the detection of an additional force. According to other advantageous but optional features of the invention, taken separately or in any technically permissible combination: said quantities comprise the actuator supply current, the control voltage of the actuator and the speed of an actuator. output member of the actuator; - Step b) of estimating said variable comprises the steps of: - b1) enter said measurements in a mathematical filter modeling the electrical and mechanical operation of the actuator and the mechanical operation of the opening; b2) implementing said filter, for example by means of a computer using a predetermined gain of said filter and matrices of a model representative of the dynamic operation of the actuator and the sash, said model being defined under form of state. the dynamic variation calculated in step d) represents the difference between the value of said variable estimated during the most recent iteration of step d) and a sliding average of said variable - the dynamic variation calculated in step d ) represents the function derived from the temporal evolution of said variable, defined for example as the difference between the value of the variable representing the additional effort estimated in step b) during the most recent iteration and the value of this variable. an estimated variable in the penultimate iteration, said difference being divided by the time interval between the last two iterations; step e) consists in comparing said dynamic variation with four thresholds determined as a function of the direction of actuation of the sash, namely opening or closing, and of the type of additional force to be detected, namely resistance to actuation of the sash presented by an obstacle or request for acceleration of the actuation of the sash formulated by a user; - The method comprises a prior step of recording a model of the kinematics of the opening and the or each threshold is determined according to this model. On the other hand, the subject of the invention is a method for controlling an electric actuator intended to open and / or close an opening, characterized in that it consists in implementing a detection method as described. above and in that it comprises a step g) consisting in transmitting the signal emitted in step f) to the actuator as a set point for stopping or accelerating the actuation of the sash.

Furthermore, the subject of the invention is a motor vehicle comprising at least one opening, such as a sliding side door, a window or a retractable roof panel, and an electric actuator for moving this door, as well as a calculator for controlling the actuator, characterized in that it further comprises means for making measurements of several quantities associated with the actuator, and in that the computer is programmed to implement a control method such as explained above. The invention will be better understood and other advantages thereof will also emerge in the light of the following description given by way of nonlimiting example and with reference to the accompanying drawings, in which - FIG. previously described, is a diagram illustrating a detection method of the prior art; FIG. 2, previously described, is a diagram illustrating another detection method of the prior art; FIG. 3, previously described, illustrates a detection device 25 of the prior art; FIG. 4, previously described, is a timing diagram illustrating the detection method employing the device of FIG. 3, previously described; - Figure 5 illustrates an opening and its actuator belonging to a vehicle according to the invention; FIG. 6 is a flowchart illustrating a detection method according to the invention for controlling the actuator of FIG. 5; FIG. 7 shows a state machine illustrating the states in which the opening of FIG. 5 can be found, as well as the transitions between these different states; and - Figure 8 is a flowchart illustrating a control method according to the invention. Figure 5 shows an opening, in this case a sliding side door P of a motor vehicle according to the invention, which can be driven by a motor M constituting a rotary type electric actuator. Alternatively, the opening can be an electric window, a retractable roof panel or any other mobile opening of a motor vehicle. The motor M rotates a drum 20, where a cable 21 can be wound or unrolled in the direction of rotation of the drum 20. The cable 21 is engaged around two pulleys 22 located on either side of the drum 20 The cable 21 drives two rollers 23 in translation parallel to an axis X and within a rail 24 adapted to guide the rollers 23. The cable 21 has in fact two strands which are each wound on the drum 20 and which are connected respectively to one and the other of the rollers 23 passing around one and the other of the pulleys 22. The door P is secured to the rollers 23 by means of a carriage 25, so that it is also driven in translation along the axis X by the motor M. The travel of the door P extends between two extreme positions, namely a closed position XF and an open position Xo. The current or instantaneous position Xp of the gate P along the X axis is therefore limited by the closed positions XF and open Xo. When the door P is actuated by the motor M, that is to say during the opening and closing of the door P, the door P has an instantaneous speed vp (t) which is variable and whose sign is arbitrarily defined as positive in the opening direction of the gate P. Conversely, the speed vp (t) is of negative sign when the gate P is being closed. To facilitate the understanding of the invention, the movement of the opening constituted by the door P is here reduced to a rectilinear translation, although it can actually consist of curvilinear translations, particularly at the end of the race, or even rotations .

It is considered that, during a normal actuation of the door P, only the motor M exerts forces on the door P. Thus, under the impetus of the motor M, the cable 21 exerts a force F21 on the door P; this effort is sometimes referred to as dragging. However, during its actuation, the door P may undergo an additional force FR which is superimposed on the actuating force F21 exerted by the cable 21 on the door P. The superimposed term here means that the additional force FR disturbs the normal actuation of the door P by the motor M, that is to say, it tends to slow or, on the contrary, to accelerate the movement of the door P. The additional force FR can therefore be resistant type or motor type.

The term "resistant" here refers to a force FR tending to slow the movement of the door P, opening or closing, and the motor term here qualifies an effort FR which tends to accelerate the movement of the door P. The motor vehicle according to the invention further comprises an electronic power circuit 3 for supplying the electric motor M, and a computer 4 for controlling the operation of the motor M and controlling the circuit 3, thus the motor M. The computer 4 is programmed to implement a method according to the invention for controlling the motor M, as described below. This control method consists in implementing a method of detecting an additional force FR, as described below. In addition, this control method comprises a step of transmitting to the motor M a stop or acceleration of the movement of the door P, according to the signals resulting from the implementation of this detection method. A method according to the invention for detecting an additional force FR is shown schematically in FIG. 6. The first step a) of this method consists in making measurements of several quantities associated with the motor M. The term associated here means that these quantities are related to the operation of the motor M. In practice, these quantities include the feed current IM, the control voltage UM and the speed of rotation wM of the motor M. The speed wM corresponds to the speed of the output motor M. In this case, it is the rotational speed of an output shaft 26 which connects the motor M to the drum 20. Alternatively, it could be a linear speed in the case where the actuator has a linear motion. The vehicle according to the invention comprises means 5 and 6, visible in Figure 5, to achieve step a), that is to say, to measure the quantities UM, IM, cOM. In this detection method, the following step b) consists of estimating, from the measurements carried out in step a), at least one variable characteristic of the forces FR and F21 undergone by the gate P. This step b) of estimation of such a variable comprises the steps b1) to b2) which are successively described below. Step b1) consists of entering the measurements UM, lm and cOM carried out in step a) in a mathematical estimator filter modeling the electrical and mechanical operation of the motor M and the mechanical operation of the door P.

The equation modeling the electrical operation of the motor M connects the supply current IM, the control voltage UM and the rotational speed cOM of the motor M: RMIM (t) = UM (t) ùKect) M (t) [E1 ] with: lm motor power supply current M; UM motor control voltage M; WM motor rotation speed M; RM internal resistance of the motor M; Ke counter-electromotive force constant of the motor M; t time.

The fundamental equation of the dynamics applied to the output shaft 26 of the motor M makes it possible to express the rotary acceleration of the motor M: JMcbM (t) = KcIM (t) ù aMCOM (t) ù R20F21 (t) [ E2] with: JM moment of inertia of the output shaft 26 of the motor M; Eon / 1 acceleration of the motor M; Kc motor torque constant M; am viscous damping in rotation; R20 radius of the drum 20; F21 force applied by the cable 21 on the door P. 9 The fundamental equation of the dynamics applied to the gate P makes it possible to express its linear acceleration: mpvp (t) = F21 (t) ù FR (t) ù apvp ( t) [E3] with: F21 force applied by the cable 21 on the door P; Additional effort exerted on the P-gate (magnitude to be estimated); mp mass of the door P; ap viscous damping in translation along the X axis; vp speed of the door P in translation along the X axis; V acceleration of the P gate in translation along the X axis.

The elongation of the cable 21 is obtained by integrating the difference of the linear velocities at the drum 20: L21 (t) = R20M (t) ùvp (t) [E4] with: L21 elongation of the cable 21. The effort in the cable 21 is directly related to its elongation: F21 (t) = K21L21 (t) [E5] with: K21 stiffness of the cable 21. The equations [E1] to [E5] can be associated in a system in the form of state: x1 (t) A1x1 (t) + B1u1 (t) [E6] YM (t) = C1x1 (t) with: x1 = [coM L21 vp] T state vector; u1 = [UM FR] T input vector; YM = [IM coM] T vector of the outputs. The matrices A1, B1 and C1 are constants and can be written as a function of the parameters associated with the equations [El] to [E5], ie Ke, Kc, JM, aM, R20, mp, ap and K21. To show the magnitude of the additional effort FR at the output of the estimation model, we adopt a constant type generator model for the force FR: FR (t) = Constant = FR (t) = 0 [E7] This choice is justified by its simplicity and the fact that there is no information to choose a generator model of another type (ramp, sinusoidal ...). By injecting the equation [E7] into the system [E6], we obtain the following model: x1 (t) Al B12 x1 (t) + BL 11 Xz (t) _ oo x2 (t) out YM (t) [ cl 0] x1 (t) x2 (t) Yo (t) = [o1] xi (t) x2 (t) with: yo = x2 = FR; B1 = [B11 B12] r, u = UM.

In vector form, these equations become: x (t) = Ax (t) + Bu (t) YM (t) = CMx (t) [E9] y0 (t) = Cox (t) with: x (t) = x1 (t)

x2 (t) A = Al B12 0 0 'B = 611 0 CM = [C1 0]; Co = [0 1]. The estimator filter associated with the model defined by the equation [E9] is then written in a classical way: x (t) = Ax (t) + Bu (t) + Ko (yM (t) ù ÿM (t)) ( t) = (A ù KoCM) x (t) + Bu (t) + KoyM (t) 9m (t) = CMx (t) [E10] [90 (t) = Cox (t) 10 (t) = Cox (t) [E8] with: estimated state vector; there. estimated output vector; K. gain of the filter estimator. The implementation of this estimator filter on a calculator requires the calculation of the gain K. This calculation is performed, for example by applying the Kalman methodology, from the constant matrices A, B and CM. These equations show: • as input: u = UM and yM = [IM coM] T; • at the output: ÿo = FR, the estimate of the additional effort FR.

Consequently, the implementation of the estimator filter [E10] in the calculator 4, during a step b2) makes it possible to estimate the additional force FR from the quantities associated with the actuation of the gate P that are the control voltage UM, the supply current lm and the rotation speed wM of the motor M.

Thus, step b) of the detection method according to the invention makes it possible to estimate, from the measurements carried out in step a), a variable characteristic of the additional force FR, which is in this case the effort FR itself. Such a variable could however be simply proportional to the additional effort FR and thus be different from it. For example, the resistive torque applied to the rotation shaft of the electric motor. The following step c) consists in recording the value of the force FR, estimated during step b), in a memory able to record several consecutive values of the additional force FR. Indeed, the force FR varies during the actuation of the gate P, from a zero or negligible value to a significant value, since it is, by definition, a sporadic effort due to the presence of an obstacle or motor force exerted by a user requesting the acceleration of the actuation. As the steps a) and b) are iterative, they make it possible to constantly estimate the value of the additional force FR. The capacity of the memory implemented in step c) is selected to record several consecutive values of the force FR, over a normal travel of the door P. To formulate a request for acceleration of the actuation of the P door, a user pulls or pushes the door P in its direction of actuation, this traction or thrust is then interpreted by the computer 4 as a request from the user and an instruction is issued to accelerate the movement of the gate P. From the values of the effort FR recorded in this memory, the calculator 4 calculates, during a step d), the dynamic variation of the force FR during the time interval which separates the first and the last iteration having given rise to the recording of values of the force FR in the memory associated with the computer 4. By dynamic means a variation of the force FR carried out during a time interval. The last iteration corresponds to the current moment, where the gate P is in the position Xp. Two calculation methods are possible to estimate the dynamic variation of the effort.

First, the first method of calculation consists of comparing the current value of the estimated effort with a sliding average of this magnitude over a time interval. This time interval can be selected relatively long, in practice between 0.5 s and 2 s, so that the variation of the force FR calculated in step d) represents a time average of the force FR. In this case, the memory periodically records, for example every 1 ms, the value of the force FR during this relatively long period of time, then it overwrites the oldest recorded values, thus realizing a so-called sliding mFR time average of the effort FR. In other words, the value of this average is recalculated each time a new value of the effort FR is estimated, that is to say after each iteration of step b). The variation of the force FR, during the relatively long time interval, is formed by the difference between the value of the effort FR (t) estimated during the most recent iteration of step b) and the sliding average mFR. We then note this variation FR ûmFR. Alternatively, the variation of the effort FR can be the derivative dFR / dt of the temporal evolution of the effort FR, that is to say the difference between the value of the variable representative of the effort FR estimated at step b) during the most recent iteration and the estimated value during a previous iteration, for example the penultimate, this difference being divided by the time interval between these last two iterations. The time interval considered is therefore a function of the time of passage to the next iteration. The derivative dFR / dt takes a null value as long as the force FR is constant, for example zero, but it varies as soon as the force FR varies.

The following step e) consists in comparing the variation of the force FR calculated in step d) with several defined thresholds as a function, in particular, of the dynamic variation calculation mode, either by the sliding average method or by the method of the derivative, the type of effort FR to detect, resistant or motor, and possibly a model of the kinematics of the P door which is pre-recorded and which can integrate for example the hard points where the movement of the door P is systematically slowed down. The following step f) consists of sending a signal indicating the detection of an additional force FR when such a threshold is exceeded. In the case of the time derivative, step e) consists of comparing this derivative dFR / dt with four thresholds S1, S2, S3 or S4 which are determined as a function of the direction of actuation of the gate P, opening or closing, and the type of effort FR that one seeks to detect, resistant or motor. The thresholds S1 to S4 are directly calibrated in Newton per second (N / s). Thus, the exceeding of the threshold S1 or the threshold S2 is due to the presence of an obstacle, therefore to a resistant force FR, respectively during the opening and closing of the door P. Exceeding the threshold S3 or S4 threshold is caused by a request for assistance from a user, so by a force FR engine, respectively during the opening and closing of the door P. In the case of the sliding average, the value of the variation FR - mFR is compared, according to step e), to four thresholds TI, T2, T3 or T4 in the same way as in the case of the derivative. The difference is that in this case the thresholds are calibrated in Newton (N). In order to determine which threshold must be compared with the estimated effort, a state machine represented in FIG. 7 is integrated in the computer 4. The states of the door P, framed in FIG. 7, are: - opening completely closed, - opening fully open, - intermediate stop: opening stopped in intermediate position, that is to say half-open, - closing in progress: opening in closing phase at normal or fast speed, - opening in progress: opening in phase of opening at normal or fast speed.

The transitions between these states, represented by arrows in FIG. 7, are: - obstacle: obstacle detection signaled, - assistance: detection of assistance request signaled, - action: request for actuation of the opening formulated by the user, - end of travel: signaled by end-of-stroke sensors during the opening phase or during the closing phase. The action transition thus makes it possible to go from the closed opening state to the current opening state. This opening is first done at normal speed, then fast speed if the user makes a request for assistance (transition assistance), that is to say, acceleration of the opening. From the current opening state, it normally goes to open open state when the end of race transition is crossed by the activation of a corresponding sensor. However, the opening movement stops to go into the intermediate state when the computer 4 sends a signal indicating the detection of an obstacle (obstacle transition). An action of the user makes it possible to go from the intermediate state or the open state to the current closed state. Closing is performed at normal speed, or at a fast speed if the user requests assistance. In case of detection of an obstacle when the closure is in progress, the movement of the door P is reversed, which leads to the current opening state. This allows the user to clear the obstacle and thus limit the consequences of the collision. If no obstacle is detected, the end of travel transition, indicated by an appropriate sensor, makes it possible to go from the current closing state to the closed opening state. The vehicle object of the invention can implement either the detection method based on a sliding average mFR, or the detection method based on the derivative dFR / dt. The vehicle object of the invention can however implement in parallel the two methods mentioned above, in order to minimize the risk of false detection. The flow diagram of FIG. 8 represents the main steps of the control-command algorithm of the motor M actuating the gate P. This algorithm comprises steps of a control method of the motor M according to the invention. After the start of the algorithm, the first test is to determine whether the user has requested the opening or closing of the door P. In the case where the opening is requested, the motor M is controlled to open the door P.

Then, the algorithm tests the end-of-race condition. If this condition is fulfilled, the gate P is in the open state and the algorithm ends. If this condition is not fulfilled, the algorithm implements a detection method as described above, starting by estimating the effort FR following step b). The derivative dFR / dt is then compared with the threshold SI to detect a possible obstacle. If the derivative dFR / dt is greater than the threshold SI, the algorithm directly controls the shutdown of the opening according to a step g) of control of the motor M. On the other hand, in the absence of an obstacle, the algorithm compares the derivative dFR / dt at the threshold S3, in order to detect a possible request for assistance or acceleration of the opening by the user. In the absence of such a request, the algorithm closes the loop by continuing to control the motor M in the direction of the opening. On the other hand, if the derivative dFR / dt is below the threshold -S3, which means that the user exerts a motor force to request the acceleration of the movement, the algorithm transmits to the motor M a fast opening instruction of the door P according to step g).

Subsequently, during this rapid opening of the gate P, the algorithm tests, as before, the end-of-travel condition and continues to estimate the effort FR permanently, according to step b), in order to detect a possible obstacle, in which case it sends to the motor M a stopping of the P door. The right half of the flowchart of Figure 8 illustrates the closing phase of the door P. The description given above of the Opening algorithm can be directly transposed to the closure algorithm. However, the closure algorithm differs from the opening algorithm by the thresholds applied to obstacle detection and assistance request, respectively S2 and S4, instead of SI and S3, and by the comparison direction. the derivative dFR / dt at these thresholds since the P gate moves in the opposite direction. On the other hand, as indicated in connection with FIG. 7, if an obstacle is detected during the closing phase, at normal speed or at a fast speed, the algorithm goes back to the opening phase to reverse the movement of the P-gate. and allow the user to clear the obstacle. A detection method according to the invention thus makes it possible to facilitate the calibration of the control-command algorithm of the motor M, since it is based on a physical model and it expresses a result in the form of a physical quantity. The force FR can indeed be given in newtons, while the detection methods of the prior art are based only on a supply current or a control voltage of the actuator. The calibration of the control-command algorithm is facilitated by the fact that the safety standards for electric window-type windows indicate a resisting force limit expressed directly in newtons, for example 100 N in the case of European standards 2000 / 4 / EC. Moreover, the invention makes it possible to integrate all the available measurements into the mathematical model of the estimating filter, whereas the detection methods of the prior art are content with a single measurement. A detection method according to the invention therefore offers an optimal accuracy for the detection of obstacles, which makes it possible to reach the best compromise between the false detection of virtual obstacles and the minimization of the damage caused by the collision with an obstacle. , for example the pinching of a finger. In other words, the invention makes it possible to reduce a pinch force without generating false detections. In addition, a vehicle according to the invention, implementing a detection method and a control method according to the invention, provides an additional benefit to users by enabling the detection of a request for assistance formulated by a user. by forcing the opening or closing movement of the opening. Such detection makes it possible to control the electric actuator the acceleration of the movement of the opening.

Claims (9)

1. Method of detecting an additional force (FR), resistant or motor, which is exerted on an opening (P) of a motor vehicle and which is superimposed on the actuating force (F21) of said opening (P) exerted by an electric actuator (M), characterized in that it comprises the iterative steps of: - a) making measurements of several quantities (IM, UM, WM) associated with the actuator (M); - b) estimate, from the measurements carried out in step a), at least one characteristic variable of said additional force (FR); - c) registering said variable in a memory able to record several consecutive values of said variable; - d) calculating, from the consecutive values recorded in said memory, the dynamic variation (FR ûmFR; dFR / dt) of the variable (FR); e) comparing said variation (FR ûmFR, dFR / dt) with at least one threshold (Ti, T2, T3, T4, Si, S2, S3, S4); and f) if said threshold (T1, T2, T3, T4, S1, S2, S3, S4) is exceeded, transmit a signal indicating the detection of an additional force (FR). 2. Method according to claim 1, characterized in that said magnitudes comprise the feed current (IM) of the actuator (M), the control voltage (UM) of the actuator (M) and the speed (WM ) an output member (26) of the actuator (M). 3. Method according to claim 1 or 2, characterized in that step b) of estimating said variable comprises the steps of: -b1) entering said measurements in a mathematical filter modeling the electrical and mechanical operation of the actuator (M) and the mechanical operation of the sash (P); and - b2) implementing said filter, for example by means of a calculator (4) by using a predetermined gain Ko of said filter and matrices (A, B, Cm, Co) of a representative model of the dynamic operation of the actuator (M) and the opening (P), said model being defined as a state. 4. Method according to one of claims 1 to 3, characterized in that the dynamic variation (FR ûmFR) calculated in step d) represents the difference between the value of said estimated variable (FR) during the most recent iteration of step d) and a running average (mFR) of said variable. 5. Method according to one of claims 1 to 3, characterized in that said dynamic variation calculated in step d) represents the derived function (dFR / dt) of the temporal evolution of said variable (FR), defined by example, as the difference between the value of the variable representing the additional effort (FR) estimated in step b) during the most recent iteration and the value of the said estimated variable in the penultimate iteration, the said difference being divided by the time interval between these last two iterations. 6. Method according to claim 4 or 5, characterized in that step e) consists of comparing said dynamic variation (FR ûmFR, dFR / dt) with four thresholds (T1, T2, T3, T4, SI, S2, S3). , S4) determined as a function of the direction of actuation of the opening (P), namely opening or closing, and of the type of additional force (FR) to be detected, namely resistance to the actuation of the sash ( P) presented by an obstacle or request for acceleration of the actuation of the sash (P) formulated by a user. 7. Method according to one of the preceding claims, characterized in that it comprises a prior step of recording a model of the kinematics of the opening (P) and in that the or each threshold (TI, T2, T3 , T4, SI, S2, S3, S4) is determined according to this model. 8. A method of controlling an electric actuator (M) intended to open and / or close an opening (P) of a motor vehicle, characterized in that it consists in implementing a detection method according to one of the preceding claims and in that it comprises a step g) of transmitting the signal emitted in step f) to the actuator (M) as an instruction to stop or accelerate the actuation of the opening (P). 9. Motor vehicle comprising at least one opening (P), such as a sliding side door, a window or a retractable roof panel, and an electric actuator (M) for moving the opening (P), and a calculator (4) for controlling the actuator (M), characterized in that it further comprises means (5, 6) for making measurements of several quantities (IM, UM, WM) associated with the actuator ( M) and that the computer (4) is programmed to implement a control method according to claim 8.