Method for detecting approach and/or contact of the hand of a user close to a door handle of a motor vehicle, associated capacitive sensor and detection module

Abstract

A method for detecting the approach and/or contact of a user's hand close to a motor vehicle door handle. The handle including: an electrode having a capacitance, and a device for measuring the variation in capacitance, a low-frequency antenna situated close to the electrode, the emissions of the low-frequency antenna bringing about disturbances in the operation of the device for measuring the variation in the capacitance. The method proposes that a new adaptive approach detection threshold be calculated, depending on an average of the previous measured variations in the capacitance, calculated over a predetermined time interval, the measured variations being weighted by a factor depending on a difference between each measured variation and the average of the variations calculated previously over a previous predetermined time interval, the approach detection being confirmed if the measured variation is higher than the new adaptive threshold.

Claims

1. A method for detecting the approach and/or contact of a user's hand close to a motor vehicle door handle, said handle comprising: an electrode having a capacitance, and a device for measuring a variation in said capacitance, a low-frequency antenna situated close to the electrode, emissions of said low-frequency antenna bringing about disturbances in the operation of the device for measuring the variation in the capacitance, the method comprising: calculating a new adaptive approach detection threshold, depending on an average of previous measured variations in said capacitance, calculating over a predetermined time interval, said measured variations being weighted by a factor depending on a difference between each measured variation and the average of the variations calculated previously over a previous predetermined time interval, the approach detection being confirmed if the measured variation in said capacitance is higher than the new adaptive threshold.

2. The detection method as claimed in claim 1, wherein the new adaptive threshold is calculated at a set frequency.

3. The detection method as claimed in claim 1, wherein the predetermined time interval and the previous predetermined time interval have equal durations.

4. A capacitive sensor for detecting the approach and/or contact of a user's hand toward a motor vehicle door handle, said sensor comprising: an electrode having a capacitance, and a device for measuring a variation in said capacitance, a low-frequency antenna situated close to the electrode, emissions of said low-frequency antenna bringing about disturbances in the operation of the device for measuring the variation in the capacitance, means for calculating a new adaptive approach detection threshold depending on an average of the measured variations in said capacitance, calculated over a time interval, said measured variations being weighted by a factor depending on a difference between each measured variation and the average of the measured variations calculated previously over a previous predetermined time interval, and means for comparing between the variation in said capacitance and the adaptive threshold in order to confirm the approach detection in order to confirm the approach detection.

5. A motor vehicle comprising a sensor as claimed in claim 4.

6. A module for detecting the approach and/or contact of a user's hand toward a motor vehicle door handle, said module comprising a capacitive sensor comprising: an electrode having a capacitance, and a device for measuring a variation in said capacitance, and a low-frequency antenna situated close to the electrode, the emissions of said low-frequency antenna bringing about disturbances in the operation of the device for measuring the variation in the capacitance, means for calculating a new adaptive approach detection threshold depending on an average of the measured variations in said capacitance, calculated over a time interval, said measured variations being weighted by a factor depending on a difference between each measured variation and the average of the measured variations calculated previously over a previous predetermined time interval, and means for comparing between the variation in said capacitance and the adaptive threshold in order to confirm the approach detection in order to confirm the approach detection.

7. The detection module as claimed in claim 6, wherein the calculating means and the comparison means take the form of software integrated into a microcontroller.

8. A motor vehicle comprising a detection module as claimed in claim 7.

9. A motor vehicle comprising a detection module as claimed in claim 6.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

(1) Other aims, features and advantages of aspects of the invention will become apparent on reading the following description, by way of nonlimiting example and on examining the appended drawings, in which:

(2) FIG. 1 (already commented upon) schematically shows a handle 10 comprising an LF antenna A and a capacitive sensor C integrated into a printed circuit 80, comprising an electrode 40 and a measurement device 60 according to the prior art,

(3) FIG. 2 (already commented upon) is a graph showing the variation in the capacitance Ce as a function of time and the detection of the approach of a hand, when the variation in capacitance moves above the adaptive threshold Nth,

(4) FIG. 3 (already commented upon) is a graph showing the variation in capacitance Ce when there is an electromagnetic disturbance I and the false approach detection, when the variation in capacitance moves above the adaptive threshold Nth,

(5) FIG. 4 schematically shows the module D comprising the capacitive sensor C and the antenna A according to an aspect of the invention,

(6) FIG. 5 is a graph showing the variation in capacitance Ce when there is an electromagnetic disturbance I and the new adaptive threshold Nth, according to an aspect of the invention, illustrating the absence of false approach detection.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

(7) As explained above, the capacitive sensor C from the prior art, when it is located close to an LF antenna A, is liable to be disturbed by the electromagnetic waves emitted by said LF antenna A. This results in an error in the measured variations in the capacitance Ce, that is to say in the signal N, in turn leading to an error in the adaptive threshold Nth. The disturbances are reflected by abrupt and strong drops in the signal N, which, when the adaptive threshold Nth is updated, therefore cause the value of said adaptive threshold Nth to drop (the adaptive threshold Nth being updated or calculated at a set frequency, for example every 3 seconds). As soon as the disturbance stops and the signal N returns to an undisturbed value (i.e.: the nominal value before the disturbance), it exceeds the adaptive threshold Nth, when the latter is updated, and an approach detection is confirmed, even though the crossing of the adaptive threshold by the signal N is in fact due to an untimely drop in the signal N due to strong electromagnetic disturbance and to an actual approach.

(8) As explained above, in the event of stray disturbances bringing about a one-off increase in the signal N, the adaptive threshold Nth also increases and the sensor C then becomes less sensitive to the approach of a user's hand.

(9) The aspect of the invention aims to rectify these problems.

(10) The example below is given for a capacitive sensor measuring the variation in capacitance Ce by way of a capacitive bridge, in which the capacitor Ce is connected in parallel with a charging capacitor Cs and a discharging capacitor discharging at a set frequency into the charging capacitor Cs.

(11) An aspect of the invention however remains applicable to any sensor C technology that measures the variation in capacitance Ce, for producing a signal N representative of the number of charges and discharges, by appropriately adjusting the value of the constant A (explained further below).

(12) The sensor C according to an aspect of the invention is shown in FIG. 4. Said sensor C comprises, as in the prior art, an electrode 40 and means 60 for measuring the variation in capacitance Ce of said electrode 40.

(13) An aspect of the invention proposes for the sensor C to furthermore comprise: means M1 for calculating a new adaptive threshold Nth, means M2 for comparing between said new adaptive threshold Nth and the variation in capacitance Ce in order to confirm the detection of the approach of the user's hand M toward the door P handle 10.

(14) The approach detection module D according to an aspect of the invention therefore comprises the sensor C such as defined above, that is to say comprising the calculating means M1 and the comparison means M2, as well as a low-frequency antenna A situated close to the electrode 40, the emissions of said low-frequency antenna A, as explained above, bringing about disturbances in the operation of the device 60 for measuring the variation in the capacitance Ce.

(15) The new adaptive threshold Nth according to an aspect of the invention is calculated from an average of the previous variations in said capacitance Ce, calculated during a predetermined duration t, said variations being weighted by a factor W depending on a difference between each variation and the previously calculated average of the variations, that is to say the second-to-last calculated average.

(16) This is illustrated in FIG. 5.

(17) ${\mathrm{Nth}}_{\mathrm{th}}^{}=S_{t\left(k-1\right)}+A$$\mathrm{Where}\text{:}$$S_{t\left(k-1\right)}=\frac{\underset{i=1}{\overset{x}{.Math.}}\left[W\left(\mathrm{Ni}\right)\mathrm{Ni}\right]}{\underset{i=1}{\overset{x}{.Math.}}W\left(\mathrm{Ni}\right)}$$W\left(\mathrm{Ni}\right)=1-\frac{1}{K}.Math.\frac{\mathrm{Ni}-S_{t\left(k-2\right)}}{S_{t\left(k-2\right)}}.Math.$$\mathrm{And}\text{:}$$S_{t\left(k-2\right)}=\frac{\underset{j=1}{\overset{z}{.Math.}}\left[W\left(\mathrm{Nj}\right)\mathrm{Nj}\right]}{\underset{j=1}{\overset{z}{.Math.}}W\left(\mathrm{Nj}\right)}$ Where: Nth.sub.tk: is the new threshold calculated for the interval tk, S.sub.t(k-1): is the weighted average of the measurements Ni during the interval t(k1), S.sub.t(k-2): is the weighted average of the measurements Nj during the interval t(k2), W(Ni): is the weighting assigned to the ith measurement Ni of the capacitance Ce, W(Nj): is the weighting assigned to the jth measurement Ni of the capacitance Ce, Ni: is the ith measurement of the capacitance Ce, Nj: is the jth measurement of the capacitance Ce, X: is the number of measurements during the interval t(k1), Z: is the number of measurements during the interval t(k2), A: is a constant greater than or equal to 0, for example between 6 and 30, K: is a constant greater than 0, for example between 2 and 50.
This is illustrated in FIG. 5.

(18) The new threshold Nth.sub.t3 that is applied during the time interval t3 is calculated from the weighted average S.sub.t2, calculated during the previous interval t2, to which a constant A is added.

(19) The weighted average S.sub.t2 is the average of the X measurements Ni of variation in the capacitance Ce during the interval t2, to which measurements a weighting W(Ni), that is to say a coefficient, is assigned, that is to say:

(20) $S_{t2}=\frac{\underset{i=1}{\overset{x}{.Math.}}\left[W\left(\mathrm{Ni}\right)\mathrm{Ni}\right]}{\underset{i=1}{\overset{x}{.Math.}}W\left(\mathrm{Ni}\right)}$

(21) The weighting coefficient W(Ni) for its part is calculated, for each measurement Ni, depending on the difference between said measurement Ni of variation in capacitance Ce and the weighted average Sots calculated at the previous instant t1, that is to say using the formula:

(22) $W\left(\mathrm{Ni}\right)=1-\frac{1}{K}.Math.\frac{\mathrm{Ni}-S_{t1}}{S_{t1}}.Math.$
Where K is a constant greater than 0.
And:

(23) $S_{t1}=\frac{\underset{i=1}{\overset{z}{.Math.}}\left[W\left(\mathrm{Nj}\right)\mathrm{Nj}\right]}{\underset{i=1}{\overset{z}{.Math.}}W\left(\mathrm{Nj}\right)}$

(24) For the sake of simplicity, it is considered that the time intervals have equal durations, t1=t2=t3, and therefore the number of measurements, X, Z in each interval are also equal to one another, that is to say X=Z.

(25) The method according to an aspect of the invention thus makes it possible to assign a high weighting for measurements having a small gap with respect to the weighted average calculated at the previous interval and a low weighting for measurements having a large gap with respect to the weighted average calculated at the previous interval.

(26) Specifically, measurements having a large gap with respect to the weighted average calculated at the previous interval are certainly measurements that are disturbed by noise. By assigning them a low weighting, the new threshold Nth is less affected by these stray measurements.

(27) In FIG. 5, the new threshold Nth.sub.t3 applied during the interval t3 is higher than the threshold Nth, calculated using the method from the prior art.

(28) During the interval t3, the variation measurements remain lower than the new threshold Nth.sub.t3, and no approach detection is confirmed, in contrast to the prior art.

(29) The new adaptive threshold Nth.sub.tk3 is applied to the measurements of variation N in capacitance Ce for an interval t3 of predetermined duration.

(30) Once this duration has expired, the method restarts for the following interval t (not shown), a new adaptive threshold Nth.sub.tk4 is calculated depending on the weighted average S.sub.t3, calculated during the previous interval t3, to which a constant A is added.

(31) The weighted average. S.sub.t3 then is the average of the Y measurements Nm of variation (not shown in FIG. 5) of the capacitance Ce during the interval t3, to which there is assigned a weighting W(Nm), which for its part is calculated for each measurement Nm, depending on the difference between said measurement Nm of variation in capacitance Ce and the weighted average S.sub.t2 calculated at the previous instant t2.

(32) In one preferred embodiment, the time intervals t(k2), t(k1) used to calculate the new adaptive threshold Nth.sub.tk3 all have identical durations, for example between 0.5 and 10 seconds, preferably t(k2)=t(k1)=3 seconds.

(33) In a second embodiment of the invention, the time intervals t(k2), t(k1) used to calculate the new adaptive threshold Nth.sub.tk3, as well as the time interval for applying the new threshold, that is to say tk, all have identical durations.

(34) In other words, the new adaptive threshold Nth is calculated at a set frequency f, being equal to:

(35) $f=\frac{1}{\left(\mathrm{tk}\right)}$

(36) Of course, said time intervals t(k2), t(k1), tk may vary in terms of duration and have different durations from one another.

(37) The method of an aspect of the invention therefore expediently makes it possible to reduce the weighting of stray measurements in the calculation of the average of the measurements, used to define the detection threshold.

(38) With the method of an aspect of the invention, false detections, triggered by the capacitance measurements moving above the detection threshold, these being caused by an abnormal drop in the measurements stemming from electromagnetic or other noise, are avoided. An aspect of the invention also makes it possible to avoid lack of detection of approach, if the disturbance creates an abnormal increase in the capacitance variation measurement signal.

(39) The detection method of an aspect of the invention therefore makes it possible to considerably improve the accuracy and the reliability of the sensor.