WO2023285251A1

WO2023285251A1 - Gesture detection method, in particular for controlling automatic opening of a motor vehicle opening element

Info

Publication number: WO2023285251A1
Application number: PCT/EP2022/068822
Authority: WO
Inventors: Ihssen Masri; Franck D'ARAUJO
Original assignee: Vitesco Technologies GmbH
Priority date: 2021-07-12
Filing date: 2022-07-07
Publication date: 2023-01-19
Also published as: FR3125134A1; FR3125134B1

Abstract

A gesture detection method using a return radiofrequency signal (SR_r(t)) that corresponds to the reflection, from a target, of an emitted radiofrequency signal (SR_em(t)), the method comprising the following steps: a/ determining a phase offset between the emitted radiofrequency signal (SR_em(t)) and the return radiofrequency signal (SR_r(t)), for a plurality of sampling times; b/ for a plurality of consecutive sampling times, calculating a current phase difference value δΦ(t), defined by: δΦ(t)=Φ(t)-Φ(t-Δt), where Φ(t) and Φ(t-Δt) are the values of the phase offset at sampling times t and t-Δt, respectively, each current phase difference value δΦ(t) reflecting a current value of a displacement of the target. The invention makes it possible to measure a movement through radar detection without a complex algorithm or expensive hardware.

Description

Title of the invention: Gesture detection method, in particular for controlling automatic opening of a motor vehicle door

Technical area

[0001] The invention relates to a gesture detection method, intended in particular to be used to remotely control the automatic opening of one or more openings of a motor vehicle, for example the automatic opening of the trunk or from a side door. The invention also relates to a system adapted to the implementation of such a method.

State of the art

[0002] Various gesture detection methods based on the transmission and reception of radio frequency signals are known in the prior art, in particular to control automatic opening of a motor vehicle door.

[0003] Throughout the text, radiofrequency signals designate frequency signals whose carrier frequency is between 3 kHz and 300 GHz. Preferably, the frequency of the carrier is between 23 GHz and 25 GHz, more preferably between 24.00 GHz and 24.50 GHz (frequency band authorized by the European regulatory authority, known as ISM for “Scientific, Industrial and Medical”) and more preferably still between 24.00 GHz and 24.25 GHz.

[0004] Known gesture detection methods are based on the transmission of a radio frequency signal towards a target, for example the foot of a user performing a predetermined gesture, and the reception of a return radio frequency signal. which corresponds to the reflection of the emitted radiofrequency signal on said target.

These methods can implement the calculation of a flight time, that is to say the time required for the signal to propagate from the transceiver to the target and from the target to to the transceiver. Such a solution has the advantage of being able to be implemented using a simple and inexpensive transceiver, emitting a continuous signal (CW, for “continuous wave”) single frequency. However, such a solution is not very suitable when the target is not very far from the transceiver, as is the case in the context of gesture detection for controlling the opening of a motor vehicle door.

[0006] Other known solutions are based on exploiting the Doppler effect, inducing a frequency shift (Doppler shift) when the signal radio frequency is reflected off a moving target.

Such solutions use, for example, signal processing based on fast Fourier transforms (FFT for “Fast Fourier Transform”). In the context of gesture detection for controlling the opening of a motor vehicle door, the gesture to be detected is however quite slow, with a speed typically comprised between 0.1 m/s and 3 m/s. This reduced speed results in a very low Doppler shift, typically less than 500 Hz for a carrier frequency of 24 GHz. This low Doppler shift induces the need for high measurement accuracy, and therefore the implementation of heavy, complex, and energy-intensive signal processing.

As a variant, the solutions based on the exploitation of the Doppler shift can use a radio frequency signal with continuous emission and frequency modulated (or FMCW, for English "Frequency Modulated Continuous Wave") In reception, the radio frequency signal emitted and the transmitted signal are mixed, to extract information relating to the Doppler shift. The end disadvantage of this solution is however that it imposes the use of broadband signals, with a bandwidth all the higher as the target to be detected is close to the transceiver. For example, to obtain an accuracy of less than 5 cm, with a target less than a meter away and a carrier frequency of 24 GHz, the bandwidth must be at least equal to 3 GHz. This bandwidth requirement results in a more complex and therefore more expensive transmission and reception system. In addition, the bandwidth requirement is incompatible with the width of an authorized frequency band, for example the ISM band.

Another solution instead uses two single-frequency signals, each at a separate frequency and with a sufficient difference between the two frequencies. We find the same limitations relating to the performance of the transmission and reception system, and to the possible incompatibility with the width of an authorized frequency band and with a low-cost system.

[0010] An object of the present invention is to provide a gesture detection method which does not have at least one of the drawbacks of the prior art.

[0011] In particular, an object of the present invention is to propose a gesture detection method suitable for short-range detection, allowing the use of inexpensive equipment, and not involving the implementation of processing of very complex signals.

Another object of the present invention is to provide a system suitable for the implementation of such a method. Disclosure of Invention

[0013] This objective is achieved with a gesture detection method, using a return radiofrequency signal which corresponds to the reflection, on a target, of an emitted radiofrequency signal, the method comprising said steps, implemented using 'at least one signal processing module: a/ from an electrical signal corresponding to the return radiofrequency signal, determination of a phase difference between the emitted radiofrequency signal and the return radiofrequency signal, said phase difference being determined for a plurality of sampling instants separated two by two by a predetermined sampling period; and b / for a plurality of first consecutive sampling instants, calculation of a current phase difference value dF(ΐ), defined by: dF(ί)=F(ΐ)-F(ΐ-Dΐ), with F(ΐ) the value of the phase shift at said first sampling instant, and with F(ΐ-Dΐ) the value of the phase shift at a second sampling instant immediately preceding said first sampling instant (when this value exists), where each current value of phase difference dF(ΐ) translates a current value of a displacement of said target.

The basic idea of the invention consists in following variations of a phase shift, where said phase shift is representative of a current distance value between the target and the transceiver. The variations in the phase shift therefore correspond to variations in said distance, so that it is possible to determine a movement carried out by the target from said variations in the phase shift. Determining the movement made by the target constitutes, in other words, gesture detection.

The method according to the invention is based on calculations of current values of a phase shift between the transmitted radiofrequency signal and the return radiofrequency signal. Such a calculation can be implemented in a manner known per se, and does not require the implementation of particularly complex algorithms. In particular, it is not necessary to calculate Fourier transforms. Thus, the method according to the invention does not implement complex, energy-intensive signal processing.

The method according to the invention does not involve the use of radio frequency signals distributed over a wide frequency band, either a frequency modulated signal or two signals each at a fixed frequency. In particular, the method according to the invention can be implemented using fixed frequency radio frequency signals. Consequently, the method according to the invention can be implemented using an inexpensive system, which does not present a particularly complex architecture in transmission and reception.

[0017] The use of a single and unique frequency in transmission also avoids any problem of incompatibility with the width of a frequency band authorized for radio frequency transmission.

[0018] The invention also does not require the use of two radio frequency transmitters, each pointed in a respective direction, and intended to identify a passage of the target in two respective places in order to deduce movement information therefrom. In the invention, only a single radiofrequency transmitter is necessary.

[0019] In addition, the invention does not involve calculating flight times, but rather phase shifts, so that the limitations linked to the precision of a flight time measurement are overcome. flight on short-range detection.

The invention thus makes it possible to measure a movement by radar detection, without a complex algorithm or expensive equipment. Preferably, the method according to the invention is implemented in a motor vehicle.

Advantageously, the method according to the invention further comprises the following step, implemented using the at least one signal processing module: c/ from each current difference value of phase dF(ΐ), calculation of a current distance variation value ôd(t), where said current distance variation value designates a displacement of the target between the first and second sampling instants.

The method according to the invention may also comprise a gesture recognition step, from a series of current values of distance variation δd(t).

[0023] Advantageously, the method further comprises a step of formulating and transmitting an instruction for unlocking and/or opening a motor vehicle door, when a predetermined gesture is recognized at the gesture recognition step.

[0024] Preferably, the emitted radiofrequency signal, at the origin of the return radiofrequency signal, has a constant amplitude and a constant frequency.

The emitted radiofrequency signal, at the origin of the return radiofrequency signal, advantageously has a central frequency comprised in a frequency band ranging from 24.00 GHz to 24.25 GHz.

In step b/, the value of the phase shift F(ΐ) at the first sampling instant, respectively the value of the phase shift F(ί-Dί) at the second sampling instant, can be determined using of two signals Io and Qo, with: - Io a phase-mixed signal, resulting from a mixing between electrical signals corresponding respectively to the return radiofrequency signal and to the emitted radiofrequency signal, and

- Qo a mixed signal in phase quadrature, resulting from a mixing between electrical signals corresponding respectively to the return radiofrequency signal, and to the emitted radiofrequency signal phase shifted by p/2.

As a variant, the value of the phase shift F(ΐ) at the first sampling instant, respectively the value of the phase shift F(ΐ-Dΐ) at the second sampling instant, can be determined using said signals Io and Qo, from which an in-phase calibration signal I _cai and a phase quadrature calibration signal Q _cai are respectively subtracted, with

- the in-phase calibration signal, I _cai , which results from a mixing between electrical signals corresponding respectively to the return radiofrequency signal and to the emitted radiofrequency signal, in the absence of a target or with a stationary target, and

- the phase quadrature calibration signal, Q _cai , which results from a mixing between electrical signals corresponding respectively to the return radiofrequency signal, and to the emitted radiofrequency signal phase-shifted by p/2, in the absence of a target or with a stationary target.

Advantageously, the sampling instants correspond to the sampling instants of an analog-digital converter.

The method according to the invention may also comprise the following steps:

- using a radiofrequency transceiver, transmission of the radiofrequency signal emitted;

- using the radiofrequency transceiver, reception of the return radiofrequency signal.

The invention also covers a system for implementing a method according to the invention, comprising:

- a radiofrequency transceiver, configured for the transmission of the transmitted radiofrequency signal and for the reception of the return radiofrequency signal; and

- at least one signal processing module, configured to implement the following steps: a/ from an electrical signal corresponding to the return radiofrequency signal, determination of a phase difference between the return radiofrequency signal and the emitted radiofrequency signal, said phase shift being determined for a plurality of sampling instants separated two by two by a predetermined sampling period; and b / for a plurality of first consecutive sampling instants, calculation of a current phase difference value dF(ΐ), defined by: dF(ί)=F(ί)-F(ί-Dΐ), with F(ί) the value of the phase shift at said first sampling instant, and with F(ΐ-Dΐ) the value of the phase shift at a second instant d sampling immediately preceding said first sampling instant, where each current value of phase difference dF(ΐ) translates a current value of a displacement of said target.

Description of figures

[0031] Other characteristics and advantages of the invention will become apparent on reading the description which follows. This is purely illustrative and must be read in conjunction with the appended drawings on which:

[0032] [Fig. 1] Figure 1 schematically illustrates a system according to the invention, suitable for the implementation of a method according to the invention;

[0033] [Fig. 2] Figure 2 schematically illustrates a method according to a first embodiment of the invention;

[0034] [Fig. 3] Figure 3 schematically illustrates a method according to a second embodiment of the invention; and

[0035] [Fig. 4] FIG. 4 is a graph showing the evolution as a function of time of distance measurements made using a method according to the invention.

Detailed description of at least one embodiment

In order to facilitate understanding of the invention, an example of a system 100 according to the invention is first described, adapted to the implementation of a method according to the invention. Preferably, but in a non-limiting way, the system 100 is configured to be mounted within a motor vehicle.

Here, the system 100 comprises the following elements:

- an electronic oscillator 101;

- a coupler 102;

- a circulator 103;

- a radio frequency transmission and reception antenna 104;

- an amplifier 105; and

- A set 106 consisting of two processing modules 16A and 16B.

The electronic oscillator 101 is configured to emit an initial electric signal SEi _nit (t), which is a frequency signal defined by a carrier at the frequency f _p , where f _p is a constant. Preferably, the initial electrical signal SEi _nit (t) is a signal of constant amplitude. Preferably, but in a non-limiting way, the frequency f _p is located in the ISM frequency band as defined in introduction. The electronic oscillator 101 may comprise a voltage-controlled oscillator, of the VCO (Voltage Controlled Oscillator) type, in which the value of the frequency f _p is a function of the amplitude of a DC voltage delivered in oscillator input. In any event, the initial electrical signal SEi _nit (t) is a fixed frequency signal.

The coupler 102 is configured to divide the initial electrical signal SE _mit (t), so as to direct part of this signal to the circulator 103 and the antenna 104, and the other part of this signal to the modules treatment 16A, 16B. The major part of the initial electric signal SEi _nit (t) is directed towards the circulator 103 and the antenna 104, and forms an electric transmission signal SE _em (t). A small part of the initial electrical signal SEi _nit (t) is directed towards the assembly 106, and forms a reference electrical signal SE _ref (t).

The circulator 103 is configured to isolate said electrical transmission signal SE _em (t), arriving from the electronic oscillator, from a return electrical signal SE _r (t), originating from a reflection on a target. Circulator 103 may comprise, for example, a Wilkinson divider or a passive circulator.

The antenna 104 is a radio frequency antenna, configured to convert the electric transmission signal SE _em (t) into a radio frequency signal transmitted SR _em (t). The transmitted radiofrequency signal SR _e (t) has the same frequency and amplitude characteristics as the electrical transmission signal SE _em (t). Here, the emitted radiofrequency signal SE _em (t) is therefore a signal of fixed frequency f _p.

The antenna 104 is further configured to receive a return radiofrequency signal SR _r (t), corresponding to the reflection, on said target (external to the system according to the invention), of the transmitted radiofrequency signal SR _em (t) . The antenna 104 is configured to convert the return radiofrequency signal SR _r (t) into a return electric signal SE _r (t), which propagates to the circulator 103.

In a variant not shown, the single antenna 104 is replaced by two neighboring antennas, respectively dedicated to signal transmission and reception.

An assembly comprising at least the electronic oscillator 101 and the radiofrequency antenna 104 forms a radiofrequency transceiver, configured for the transmission of the transmitted radiofrequency signal SR _e (t) and for the reception of the return radiofrequency signal SR _r (t). This assembly is configured here to transmit a single-frequency signal at the frequency f _p , so that it does not require the implementation of complex architectures and/or hardware.

In a simple variant, the initial electrical signal SEi _nit (t) and the transmitted radio frequency signal SR _em (t) are both transmitted continuously. As a variant, the emitted radiofrequency signal SR _e (t) is emitted at regular intervals, like a pulsed signal (but with a constant amplitude for the duration of the emission so that we always speak of a signal of constant amplitude). The transmission time intervals can be controlled via a switch placed between the oscillator 101 and the antenna 104, or directly at the level of the oscillator 101.

At circulator 103, return electric signal SE _r (t) is directed to amplifier 105. Amplifier 105 is preferably a low-noise amplifier, with a gain of between 20 dB and 30 dB. It makes it possible to amplify the return electric signal SE _r (t), which initially has a reduced amplitude since only a small part of the emitted radiofrequency signal SR _em (t) reaches the target and then returns to the antenna. 104. The signal at the output of the amplifier is called the amplified return electric signal, SEr_am(t).

The first processing module 16 A is configured to receive as input the amplified return electrical signal, SE _{r am} (t), coming from the amplifier 105, as well as the reference electrical signal SE _ref (t) having been taken from coupler 102.

The first processing module 16A is configured to implement signal processing using these two electrical signals, in order to obtain a phase-mixed signal, Io(t), and a quadrature-phase mixed signal, Qo (t). The signals Io(t) and Qo(t) are obtained by mixing the amplified return electric signal SE _r-am (t) and the reference electric signal SE _ref (t), in a manner known per se. The first processing module 16A therefore comprises two separators, two mixers, and a phase-shifting element, configured together to obtain the signals Io(t) and Qo(t).

The Io(t) and Qo(t) signals are then sent to the second processing module 16B. If necessary, the system 100 according to the invention may comprise a band-pass filter, not shown, arranged between the output of the first processing module 16A and the input of the second processing module 16B, and configured to filter the noise on the signals Io(t) and Qo(t) by suppressing their very rapid variations.

The second processing module 16B comprises at least one processor, and if necessary one or more memories. It is configured to implement signal processing using these two signals Io(t) and Qo(t), so as to determine phase shift values <E>(t) between the transmitted radiofrequency signal SR _em (t) and the return radiofrequency signal SR _r (t), for a plurality of sampling instants “t”.

In practice, the phase difference between the reference electric signal SE _ref (t) and the amplified return electric signal SE _{r am} (t) is calculated, this phase difference being equal to the phase difference F(ΐ) between the transmitted radiofrequency signal SR _e (t) and the signal return radio frequency SR _r (t). If necessary, the carrier of the electrical reference signal SE _ref (t) is simply used.

The sampling instants "t" preferably correspond to the sampling instants of an analog-digital converter in the system 100 according to the invention, in particular an analog-digital converter receiving as input the return electrical signal SE _r (t), or an analog-digital converter receiving the amplified return electrical signal SE _r am(t) as input, or even an analog-digital converter receiving the signals Io(t) and Qo(t) as input.

When the transmitted radiofrequency signal SR _e (t) is transmitted at regular time intervals, like a pulsed signal, only the sampling instants associated with the signal transmission are considered.

The second processing module 16B is further configured to calculate current phase difference values dF(ΐ), where dF(ΐ)=F(ΐ) - F(ΐ-Dΐ), and where "t- Dΐ” and “t” are two directly successive sampling instants.

The second processing module 16B is configured to output a series of current phase difference values dF(ΐ), for a plurality of successive sampling instants “t”. Each current value of phase difference dF(ΐ) translates a current value of a displacement of the target, between the instants "t-Dΐ" and "t". The series of current phase difference values dF(ΐ) therefore represents the evolution, as a function of time, of a position of the target, in other words a movement of the target, or still otherwise said a gesture performed by the target .

Here, but not limited to, the first processing module 16A is integrated on a chip also receiving elements such as the coupler 102, the circulator 103, and the amplifier 105. The second processing module 16B is here an annex module to said chip.

Here, the system 100 according to the invention is configured to be on board a motor vehicle. The gesture detection implemented is intended to control automatic opening of a door of said vehicle, in particular automatic opening of the trunk lid. The detected gesture is a gesture of a user located outside the vehicle, preferably a gesture of the foot, where the foot forms the target mentioned above. As a variant, the opening of the vehicle can be a side door, or any other opening of a vehicle. Likewise, the gesture detected can be a gesture of the hand, or any other part of the human body then forming the target.

Next, with reference to FIG. 2, a method according to a first embodiment of the invention is described. The method here comprises a first step 210, of transmitting the emitted radio frequency signal SR _e (t), using the transceiver as described with reference to FIG. 1.

The method then comprises a second step 220, of receiving the return radiofrequency signal SR _r (t), using the transceiver as described with reference to FIG. 1. The return radiofrequency signal SR _r (t) corresponds to the reflection, on a target located at a distance from the transceiver, of the transmitted radiofrequency signal SR _e (t).

These two steps 210 and 220 are not necessarily part of the method according to the invention, although the return radiofrequency signal SR _r (t) is necessary for the implementation of the method according to the invention.

The method according to the invention then comprises a step 230 of determining current values of a phase shift F(ΐ) between the transmitted radiofrequency signal SR _e (t) and the return radiofrequency signal SR _r (t). As specified above, the values of the phase shift F(ΐ) are determined for a plurality of sampling instants “t”, which preferably correspond to the sampling instants of an optical-digital anal converter. Said sampling instants are separated two by two by a time interval Dΐ, called the sampling period. In practice, and as detailed previously, a phase shift is calculated between electrical signals, corresponding respectively to the return radiofrequency signal SR _r (t) and to the transmitted radiofrequency signal SR _e (t).

The method then includes a step 240 of calculating current values of a phase difference dF(ΐ), for a plurality of directly successive sampling instants t, with: dF(ΐ)=F(ΐ) - F(ΐ-Dΐ), with:

F(ΐ) the value of the phase shift associated with the sampling instant t, F(ΐ-Dΐ) the value of the phase shift associated with a sampling instant t immediately preceding the instant t, and

Dΐ is the duration separating two directly successive sampling instants.

[0064] Each current value of phase difference dF(ΐ) translates a current value of a displacement of the target, between the instants “t-Dΐ” and “t”. The series thus obtained of current values of phase difference dF(ί) therefore reflects the evolution, as a function of time, of the position of the target.

Preferably, step 230 comprises the following sub-steps, implemented for each sampling instant t:

- mixing of electrical signals corresponding respectively to the return radiofrequency signal SR _r (t) and to the transmitted radiofrequency signal SR _e (t), for obtaining an in-phase mixed signal, Io(t); and

- mixing of electrical signals corresponding respectively to the return radiofrequency signal SR _r (t), and to the transmitted radiofrequency signal SR _e (t) phase-shifted by p/2, to obtain a mixed signal in phase quadrature, Qo(t).

In practice, the amplified return electric signal SE _{r am} (t) can be mixed with the reference electric signal SE _ref (t) (if necessary phase-shifted by p/2) or even with the sole carrier of the electric signal of reference SE _ref (t) (if necessary phase-shifted by p/2).

The current value of the phase shift F(ΐ) can then be given by:

<D(t)=arctan (Qo(t)/Io(t))

(the sign “=” reflects the fact that noise is not taken into account here).

Advantageously, step 230 for determining current values of the phase shift F(ΐ) includes a sub-step for applying calibration data I _cai and Q _cai .

I _rai is obtained by mixing electrical signals corresponding respectively to a return radiofrequency signal SR _r -o and to an emitted radiofrequency signal SR _e - o, obtained under calibration conditions, that is to say in l no target or with a stationary target.

Q _cai is obtained by mixing electrical signals corresponding respectively to the return radiofrequency signal SR _r -o, and to the transmitted radiofrequency signal SR _e -o phase shifted by p/2.

The data I _cai and Q _cai are determined during a preliminary calibration step, and stored in a memory of the system 100 according to the invention. They quantify a measurement noise associated with the system 100 according to the invention and with the surrounding environment, under normal conditions of use.

These calibration data are used to calculate corrected values I _CO r(t) and Q¥r(t) defined by:

Icor(t) ^— Io(t) - Ical j and Qcor(t) ⁼ Qo(t) - Qcal.

The current value of the phase shift F(ΐ) can then be given by:

®(t)=arctan (Q _CO r(t)/Icor(t)).

The subtraction of the calibration data I _cai and Q _cai , from the raw values Io and Qo, makes it possible to reduce distortion of the phase shift measurement and to perform noise filtering.

We then describe, with reference to Figure 3, a method according to a second embodiment of the invention, which will only be described for its differences relative to the method of Figure 2.

As in the method of Figure 2, the method of Figure 3 comprises the following steps:

- the step 310 of transmitting the transmitted radio frequency signal SR _em (t); then

- step 320 of receiving the return radio frequency signal SR _r (t); then - the step 330 of determining current values of the phase shift F(ΐ); then

- step 340 for calculating current values of the phase difference dF(ί).

Here, the method further comprises a step 350 of calculating, for each of said current values of the phase difference dF(ΐ), a corresponding current value of distance variation, δd(t). The link between dF(ΐ) and ôd(t) is given by: dF(ΐ)=2*p* ôd(t)/Xp with k=c/f _p , where c is the speed of light in vacuum and f _p the frequency of the carrier of the transmitted radiofrequency signal SR _e (t).

Said current distance variation value ôd(t) denotes a variation in the distance between the target and the transceiver used for steps 310 and 320, between times “t” and “ΐ-Dΐ”. In other words, it is the displacement of the target, between the instants "t" and "ΐ-Dΐ". The monitoring of the current values of δd(t) thus defines a movement carried out by the target for a predetermined duration.

In Figure 4, by way of illustration, there is shown a graph showing the evolution, as a function of time, of a position of the target. The y-axis is time t, expressed in seconds. The abscissa axis is a distance d, expressed in meters.

Arbitrarily, it is considered that at the initial instant t=0, the target is positioned at an origin point defined by a zero distance value.

The graph in Figure 4 was obtained using a system as shown in Figure 1, with the target of a user's foot. The user performs with the foot, at regular intervals, a pendulum movement back and forth (or kick, or "kick", in English).

The distance variations observed between t=0 and t=1.2s correspond to noise. Between t=1.2s and t=1.5s, a movement of the foot is observed in a first direction. Between times t=1.5s and t=1.9s, a movement of the foot in the opposite direction is observed. Between the instants t=1.9s and t=2.2s, a movement of the foot is again observed in the first direction, to return to its initial position. This corresponds well to a pendulum movement in which the foot is swung backwards, then forwards, then returns to its position of support on the ground. This movement is well observed several times over time. FIG. 4 therefore shows that the method and the device according to the invention effectively make it possible to reliably measure a movement of a target, in particular a foot performing a pendulum movement back and forth.

If necessary, a preliminary calibration step can be implemented to then make it possible to discriminate between an approaching movement and a retreating movement. [0084] Advantageously, the method may further comprise a noise filtering step, not shown, in particular using a low-pass filter, to suppress the very rapid variations of the function dF(ΐ) and/or of the function od(t).

Throughout the text, gesture detection simply consists in measuring a movement of the target, either by tracking phase shift values, or by tracking shifts in position.

Here, the method according to the invention further comprises an optional step 360 of gesture recognition. During this step 360, the movement of the target is analyzed to be classified in at least one of several predetermined categories. This step can implement at least one comparison with predetermined curve shapes. In addition or as a variant, this step can implement at least one comparison with predetermined curve characteristics, in particular movement directions and/or threshold values.

[0087] For example, step 360 may involve determining whether or not the target performs a pendulum movement, by seeking whether or not a movement consisting of the following elementary movements is found: movement in a first direction and greater than a first threshold, followed by a displacement in the opposite direction and greater than a second threshold, followed by a new displacement in the first direction and greater than a third threshold. The threshold conditions make it possible to take into account only displacements of sufficient amplitude, in order to limit the errors due to noise.

Here, the method according to the invention further comprises an optional step 370 of formulating an instruction CO for unlocking and/or opening a vehicle door, when a predetermined gesture has been recognized in step 360. The instruction CO is for example an instruction for opening the trunk of a motor vehicle. Said predetermined gesture is for example the movement of the foot as described above. As a variant, the instruction CO can be an instruction for opening a side door, or any other opening of a vehicle. Likewise, the gesture detected can be a gesture of the hand, or any other part of the human body then forming the target. In any event, the CO setpoint is transmitted to a door opening control system in the motor vehicle.

Similarly, the invention also covers a system according to the invention, configured to implement at least one of the steps 350, 360 and 370 of the method of FIG. 3 (using the processing module 16B, or at least one additional processing module comprising at least one processor and, where applicable, one or more memories).

The invention is not limited to the examples detailed above, and numerous other system and method variants according to the invention can be implemented without departing from the scope of the invention, for example with radiofrequency signals centered on other frequency values.

[0091] Preferably, but in a non-limiting way, the method and the device according to the invention are intended to be implemented in a motor vehicle, with the aim of detecting a gesture made by a user located outside the vehicle. and wishing to access said vehicle. This may be to control an automatic opening of the trunk, or another opening such as a door. The gesture controlling the opening is for example a gesture of the foot, the foot then forms the target from which a movement is detected. Alternatively, the target may consist of a hand, an elbow, or any other part of the user's body. Likewise, the invention is not limited to a particular type of movement to be detected.

As a variant, the method and the device according to the invention can be implemented to carry out gesture detection in a context other than the automobile, for example to remotely control household appliances, to open a door of a building, etc.

Claims

[Claim 1] Gesture detection method using a return radiofrequency signal (SR _r (t)) which corresponds to the reflection, on a target, of a transmitted radiofrequency signal (SR _e (t)), the method comprising the said step, implemented using at least one signal processing module (16A, 16B): a/ from an electrical signal corresponding to the return radiofrequency signal, determination (230; 330) of a phase difference between the transmitted radiofrequency signal (SR _e (t)) and the return radiofrequency signal (SR _r (t)), said phase shift being determined for a plurality of sampling instants separated two by two by a predetermined sampling period; wherein the method further comprises the following step, implemented using the at least one signal processing module (16A, 16B): b / for a plurality of first consecutive sampling instants, calculating (240; 340) of a current phase difference value dF(ΐ), defined by: dF(ί)=F(ΐ)-F(ΐ-Dΐ), with F(ΐ) the value of the phase shift at said first sampling instant, and

F(ΐ-Dΐ) the value of the phase shift at a second sampling instant immediately preceding said first sampling instant, where each current value of phase difference dF(ΐ) translates a current value of a displacement of said target, the value of the phase shift F(ΐ) at the first sampling instant, respectively the value of the phase shift F(ί-Dί) at the second sampling instant, being determined using two signals Io and Qo, with

- Io a signal mixed in phase, resulting from a mixing between electrical signals corresponding respectively to the return radiofrequency signal (SR _r (t)) and to the transmitted radiofrequency signal (SR _e (t)), and

- Qo a mixed signal in phase quadrature, resulting from a mixing between electrical signals corresponding respectively to the return radiofrequency signal (SR _r (t)), and to the transmitted radiofrequency signal (SR _e (t)) phase shifted by p/2 , characterized in that the value of the phase shift F(ΐ) at the first sampling instant, respectively the value of the phase shift F(ΐ-Dΐ) at the second sampling instant, is determined using the said signals Io and Qo, from which an in-phase calibration signal I _cai and a quadrature-phase calibration signal are respectively subtracted Qcai, with

- the in-phase calibration signal, I _cai , which results from a mixing between electrical signals corresponding respectively to the return radiofrequency signal (SR _r (t)) and to the emitted radiofrequency signal (SR _e (t)), in the absence of target or with a stationary target, and

- the phase quadrature calibration signal, Q _cai , which results from a mixing between electrical signals corresponding respectively to the return radiofrequency signal (SR _r (t)), and to the emitted radiofrequency signal (SR _e (t)) out of phase of p/2, in the absence of a target or with a stationary target.

[Claim 2] Method according to claim 1, characterized in that it further comprises the following step, implemented using the at least one signal processing module: c/ from each value current phase difference dF(ΐ), calculation (350) of a current distance variation value ôd(t), where said current distance variation value designates a displacement of the target between the first and second instants of sampling.

[Claim 3] Method according to claim 2, characterized in that it further comprises a step (360) of gesture recognition, from a series of current values of distance variation ôd(t).

[Claim 4] Method according to claim 3, characterized in that it further comprises a step (370) of formulating and transmitting an instruction for unlocking and/or opening a motor vehicle door, when a gesture predetermined is recognized at step (360) of gesture recognition.

[Claim 5] Method according to any one of Claims 1 to 4, characterized in that the transmitted radiofrequency signal (SR _e (t)), at the origin of the return radiofrequency signal (SR _r (t)), has a constant amplitude and constant frequency.

[Claim 6] Method according to any one of Claims 1 to 5, characterized in that the transmitted radiofrequency signal (SR _e (t)), at the origin of the return radiofrequency signal (SR _r (t)), has a center frequency within a frequency band from 24.00 GHz to 24.25 GHz.

[Claim 7] Method according to any one of Claims 1 to 6, characterized in that the instants of sampling correspond to the instants of sampling of an anal ogical-digital converter.

[Claim 8] Process according to any one of Claims 1 to 7, characterized in that it also comprises the following steps:

- using a radiofrequency transceiver (101, 104), transmission of the transmitted radiofrequency signal;

- using the radiofrequency transceiver (101, 104), reception of the return radiofrequency signal.

[Claim 9] A system (100) for carrying out a method according to claim 8, comprising:

- a radiofrequency transceiver configured for the transmission of the transmitted radiofrequency signal for the reception of the return radiofrequency signal (SR _r (t))

- at least one signal processing module

configured to implement the following steps: a/ from an electrical signal corresponding to the return radiofrequency signal, determination (230; 330) of a phase shift between the return radiofrequency signal and the emitted radiofrequency signal, said phase shift being determined for a plurality of sampling instants separated two by two by a predetermined sampling period; and b / for a plurality of first consecutive sampling instants, calculation (240; 340) of a current phase difference value dF(ί), defined by: dF(ΐ)=F(ί)-F(ΐ -Dΐ), with F(ί) the value of the phase shift at said first sampling instant, and with F(ΐ-Dΐ) the value of the phase shift at a second sampling instant immediately preceding said first sampling instant, where each current phase difference value dF(ΐ) translates a current value of a displacement of said target, where the value of the phase shift F(ΐ) at the first sampling instant, respectively the value of the phase shift F(ΐ-Dΐ) at second sampling instant, is determined using two signals Io and Qo, with

- Io a signal mixed in phase, resulting from a mixing between electrical signals corresponding respectively to the return radiofrequency signal (SR _r (t)) and to the transmitted radiofrequency signal (SR _em (t)), and

- Qo a mixed signal in phase quadrature, resulting from a mixing between electrical signals corresponding respectively to the return radiofrequency signal (SR _r (t)), and to the transmitted radiofrequency signal (SR _e (t)) phase shifted by p/2 , and where the value of the phase shift F(ΐ) at the first sampling instant, respectively the value of the phase shift F(ΐ-Dΐ) at the second sampling instant, is determined using said signals Io and Qo, from which one subtracts respectively an in-phase calibration signal Icai and a Q phase quadrature calibration signal _cai , with