System for detecting the presence of life inside a motor vehicle

Abstract

The invention relates to a system (1) for detecting the presence of life inside a motor vehicle (V) by means of a radar (10) and an accelerometer (40), the system being configured to implement the following steps: - processing, in particular by means of a presence-detection algorithm, data from the radar (10) in order to deliver a result regarding a presence of life in the vehicle, - processing data from the accelerometer (40) in order to deliver a result regarding a movement of the vehicle, - processing, in particular by means of a correlation algorithm, the data from the radar (10) and the data from the accelerometer (40) in order to establish a presence of correlation or an absence of correlation between at least one feature derived from the data from the radar (10) and at least one feature derived from the data from the accelerometer (40), the presence of correlation being representative of a state of actual movement of the radar (10), - in the event of presence of correlation, ignoring the result regarding a presence of life in the vehicle obtained by processing the data from the radar (10).

Description

[0001] DESCRIPTION

[0002] Title: Vehicle Life Presence Detection System

[0003] [1] The invention relates to a system for detecting the presence of life within a motor vehicle.

[0004] [2] It is known to detect the presence of life, particularly a person inside a motor vehicle, followed by an alert if a person is left inside the vehicle. Such detection and alert systems enhance passenger safety, particularly by preventing the danger of leaving a child inside the vehicle. Such detection can, for example, utilize radar located inside the vehicle.

[0005] [3] In particular, there is a need to prevent false alarms. Such a false alarm can occur when a person outside the vehicle causes vibrations to the vehicle, for example by leaning on it. The radar then erroneously detects the presence of life inside the vehicle when no one is actually inside.

[0006] [4] A false alarm can also occur when strong wind creates vibrations inside the vehicle or when someone or something comes against the vehicle, causing the radar to move.

[0007] [5] The invention aims in particular to remedy these difficulties.

[0008] [6] The invention thus relates to a system for detecting the presence of life within a motor vehicle by means of a radar and an accelerometer, the system being configured to implement the following steps:

[0009] - to process radar data, notably using a presence detection algorithm, to deliver a result on the presence of life in the vehicle,

[0010] - to process accelerometer data to deliver a result on vehicle movement,

[0011] - to process, in particular by a correlation algorithm, radar data and accelerometer data to establish the presence or absence of correlation between at least one characteristic from the radar data and at least one characteristic from the accelerometer data, the presence of correlation being representative of a state of inherent radar motion,

[0012] - If a correlation is present, ignore the result regarding the presence of life in the vehicle obtained by processing the radar data. [7] Advantageously, if no correlation is present, the system is configured to confirm the result regarding the presence of life in the vehicle obtained by processing the radar data.

[0013] [8] The accelerometer detects movements specific to the radar. These radar movements can, for example, be the result of movements such as jolts experienced by the vehicle due to an external action (e.g., strong winds, a person leaning on the vehicle). The radar alone cannot determine whether the jolt is due to an external action or to an action within the vehicle's passenger compartment (e.g., a child moving in the back seat).

[0014] [9] Thus, if the radar moves and the accelerometer detects these movements, it can be deduced that the movements detected by the radar most likely originate from an action external to the vehicle, and the result indicating the presence of life in the vehicle obtained through radar data processing is ignored. This avoids generating false positives, which are alerts for the presence of life when no one is actually in the vehicle. The use of the accelerometer allows, in a way, the filtering of the radar's own movements due to vibrations, for example. When the vehicle is jolted, the radar and accelerometer responses occur simultaneously or almost simultaneously.

[0015]

[0010] According to one aspect of the invention, the correlation algorithm is configured to detect, over a given time window or at specific points, a similarity between the signals captured by the radar and the signals captured by the accelerometer. A time window may, for example, be between 5 and 15 seconds, particularly between 7 and 12 seconds, and more particularly less than or equal to 10 seconds.

[0016]

[0011] This allows us to deduce that we are more likely in a situation of radar vibrations than a breathing pattern of a person, for example a child or an adult, in the passenger compartment.

[0017]

[0012] The invention thus makes it possible to make the life presence detection system more reliable.

[0018]

[0013] According to one aspect of the invention, said at least one characteristic derived from the radar data for establishing the correlation is measured as a function of time, for example, so as to define a curve. In particular, said at least one characteristic derived from the radar data for establishing the correlation is at least one characteristic of a maximum power curve detected by the radar as a function of time.

[0014] Other characteristics derived from the radar data can be used to establish the correlation, possibly in addition to the characteristic of the maximum power curve. Among these other characteristics that can be used to establish the correlation, examples include the growth rate or a positive or negative derivative. Generally speaking, many parameters can be considered to determine the correlation of two curves.

[0019]

[0015] According to one aspect of the invention, said at least one feature derived from the accelerometer data to establish the correlation is at least one feature of a jerk curve as a function of time. For example, one could use a gradient of the acceleration measured by the accelerometer, or use the accelerations along the three axes (X, Y, Z) of a three-dimensional orthogonal coordinate system representing space.

[0020]

[0016] According to one aspect of the invention, the system is configured to detect a correlation:

[0021] - between one or more peaks on the respective curves obtained by the radar and by the accelerometer.

[0022]

[0017] According to another aspect of the invention, the system is configured to detect a correlation:

[0023] - between the amplitude of the peaks on the respective curves obtained by the radar and by the accelerometer.

[0024]

[0018] According to another aspect of the invention, the system is configured to detect a correlation:

[0025] - between the time interval between two successive peaks on the respective curves obtained by the radar and by the accelerometer.

[0026]

[0019] According to another aspect of the invention, the system is configured to detect a correlation:

[0027] - between the shape of the radar data curve and the shape of the accelerometer data curve.

[0028]

[0020] This peak or these peaks of the radar and accelerometer curves are then the characteristics to be compared in the sense of the invention.

[0029]

[0021] For example, a correlation is determined between the power peaks of the radar curve and the "jerk" peaks of the accelerometer curve.

[0022] Generally, characteristics are chosen that exhibit a good correlation with each other when the radar undergoes its own movements, and the accelerometer measures these movements.

[0030]

[0023] The determination of the correlation is carried out over a time window, or at a specific point in time.

[0031]

[0024] According to one aspect of the invention, the correlation algorithm is configured to use at least one similarity threshold such that above a similarity threshold, a correlation is deemed present, and below this similarity threshold, an absence of correlation is deemed present.

[0032]

[0025] In particular, the similarity threshold may be a fixed threshold that allows comparison to this similarity threshold, a difference between two values ​​of the characteristics of the radar and the accelerometer, or a percentage of one of the values ​​relative to the other value. For example, the percentage of one of the values ​​relative to the other value is greater than or equal to 70%, preferably greater than or equal to 80%, more preferably greater than or equal to 90%.

[0033]

[0026] According to one aspect of the invention, the determination of the correlation is carried out after filtering the radar data and / or filtering the accelerometer data, for example to retain only the points above a predetermined filtering threshold (for example a filtering threshold equal to 60%, 70%, or even 80%).

[0034]

[0027] According to one aspect of the invention, the system is configured to detect any breathing (for example, of a child) occurring in the passenger compartment of the vehicle, to deliver the result on the presence of life in the vehicle.

[0035]

[0028] According to one aspect of the invention, the accelerometer is configured to measure an acceleration along the three axes (X, Y, Z) of a three-dimensional orthogonal frame forming space.

[0036]

[0029] The invention further relates to a method for detecting the presence of life inside a motor vehicle using a radar and an accelerometer, the method comprising the following steps:

[0037] - to process radar data, notably using a presence detection algorithm, to deliver a result on the presence of life in the vehicle,

[0038] - to process accelerometer data to deliver a result on vehicle movement,

[0039] - to process, in particular by a correlation algorithm, radar data and accelerometer data to establish the presence or absence of correlation between at least one characteristic from the radar data and at least one characteristic from the accelerometer data, the presence of correlation being representative of a state of inherent radar motion,

[0040] - If a correlation is present, ignore the result regarding the presence of life in the vehicle obtained through radar data processing.

[0041]

[0030] Other features, details and advantages of the invention will become clearer upon reading the following description on the one hand, and several illustrative and non-limiting examples of embodiments given with reference to the accompanying schematic drawings on the other hand, in which:

[0042]

[0031] [Fig 1] Figure 1 is a schematic view of a life presence detection system within a motor vehicle according to an example of an embodiment of the invention;

[0043]

[0032] [Fig. 2] Figure 2 is a block diagram representing the steps of determining the presence of life by the system of Figure 1;

[0044]

[0033] [Fig. 3] Figure 3 illustrates a correlation method used in the invention.

[0045]

[0034] The features, variants, and different embodiments of the invention can be combined in various ways, provided they are not incompatible or mutually exclusive. In particular, variants of the invention may be conceived comprising only a selection of features, described hereafter in isolation from the other described features, if this selection of features is sufficient to confer a technical advantage and / or to differentiate the invention from the prior art.

[0046]

[0035] Figure 1 shows a system 1 for detecting the presence of life within a motor vehicle V according to an example of an embodiment of the invention, to alert on the presence of an occupant 30 (for example a forgotten child) within the vehicle V. The system 1 comprises a presence sensor, here a radar 10, and a computer 20 connected to the radar 10.

[0047]

[0036] The radar 10, which is a radar wave transmission and reception device, is configured to determine the presence of an occupant 30 within the vehicle V.

[0048]

[0037] The radar 10 is placed in a central area of ​​the vehicle V, here in the center of the roof 2 of the vehicle V as shown in figure 1, the radar 10 can then advantageously cover all the seats of the vehicle V.

[0049]

[0038] Of course, in the invention, several radars can be used by being distributed in different locations in the passenger compartment of the vehicle V.

[0050]

[0039] The radar 10 here includes at least one antenna (not shown) designed to emit the electromagnetic wave and at least one sensor (not shown) designed to receive a reflected electromagnetic wave after reflection of the electromagnetic wave, and in particular after reflection on a person 30 inside the vehicle V.

[0051]

[0040] The radar 10 uses millimeter electromagnetic waves. The radar is, for example, a Doppler radar using sustained or continuous waves.

[0052]

[0041] The control unit 20 comprises at least one memory and at least one processor. The control unit 20 may be, for example, the electronic control unit (ECU) of vehicle V. The control unit 20 may also be a control unit dedicated to system 1. Instructions enabling the determination of the presence of life status are stored in the memory and implemented by the processor. When implemented, these instructions enable the execution of the process described later.

[0053]

[0042] The life presence detection system 1 further includes an accelerometer 40 fixed to the vehicle V so as to undergo the same movements as the radar 10 when the vehicle V is subjected to movements or vibrations.

[0054]

[0043] Preferably, the accelerometer 40 is located near the radar 10, or even mounted on or in the radar 10. Advantageously, the proximity of the accelerometer 40 to the radar 10 makes it possible to determine whether the radar has made a movement.

[0055]

[0044] The accelerometer 40 is thus configured to detect movements specific to the radar 10. These movements of the radar 10 could, for example, be the result of movements such as jolts experienced by the vehicle V due to an external action (for example, strong winds, a person leaning on the vehicle, etc.). The radar 10 alone cannot determine whether the jolt is due to an external action or to an action within the passenger compartment of the vehicle V (for example, a child sitting on the back seat moving).

[0056]

[0045] Thus, if the radar 10 moves and the accelerometer 40 detects these movements, it can be deduced that the movements detected by the radar 10 originate, with a high probability, from an external action, and the result on the presence of life in the vehicle V obtained by processing the data from the radar 10 is ignored.

[0057]

[0046] The computer 20 is connected to the accelerometer 40 and the radar 10, in particular to control the emission of the electromagnetic wave.

[0058]

[0047] As illustrated in Figure 2, system 1 is configured to implement the following steps:

[0059] - process radar 10 data using an ALP presence detection algorithm to deliver a result on the presence of life in the vehicle (step S1), - process accelerometer 40 data to deliver a result on vehicle movement (step S2),

[0060] - process, by an ALC correlation algorithm, data from radar 10 and data from accelerometer 40 to establish a presence or absence of correlation between at least one CR characteristic from radar 10 data and at least one CA characteristic from accelerometer 40 data, the presence of correlation being representative of a state of proper motion of radar 10 (step S3).

[0061]

[0048] In the event of a correlation, system 1 is configured to ignore the RD result on the presence of life in the vehicle obtained by processing the radar data 10 (step S4).

[0062]

[0049] In the absence of correlation, system 1 is configured to confirm the RD result on the presence of life in the vehicle obtained by processing the radar data 10 (step S5).

[0063]

[0050] System 1 is configured to detect, using the ALP presence detection algorithm, any breathing (for example, of a child) occurring inside the vehicle's passenger compartment, and to provide a result indicating the presence of life in the vehicle. For example, the detection of life using radar 10 relies on statistical measurements to extract dispersion information from radar data (in particular, pixels of the radar image). These measurements are then, for example, compared to a data library to conclude, following a classification process, whether a person is present in the passenger compartment.

[0064]

[0051] The ALC correlation algorithm is configured to detect, over a given time window or at specific points, a similarity between the signals captured by the radar 10 and the signals captured by the accelerometer 40

[0065]

[0052] A time window may for example be between 5 and 15s, particularly between 7s and 12s, more particularly be less than or equal to 10s.

[0066]

[0053] This allows us to deduce that we are more likely in a situation of inherent vibrations of the radar 10 than of a breathing pattern of a person, for example a child or a person with a disability, in the passenger compartment.

[0067]

[0054] The invention thus makes it possible to make the life presence detection system more reliable.

[0068]

[0055] Said at least one CR characteristic from the radar 10 data to establish the correlation is measured as a function of time, for example so as to define a curve, here a C1 curve of maximum power (see figure 3), in ordinates, detected by the radar 10 as a function of time, in abscissa (in seconds).

[0069]

[0056] Other features from the radar data 10 could be used to establish the correlation, possibly in addition to the maximum power curve characteristic. Other features from the radar data 10 include, for example, the number of points in a point cloud, the dispersion of points in a point cloud, the velocity of points in a point cloud, the growth rate, a positive or negative derivative, etc. Any features that allow for finding a correlation relationship can, of course, be used.

[0070]

[0057] The CA characteristic derived from the accelerometer 40 data to establish the correlation is at least one characteristic of a jerk curve as a function of time. This characteristic is illustrated on a curve C2, in Figure 3.

[0071]

[0058] On this curve C2, the ordinate shows the value of the jolt, which is the sum of the derivatives of the acceleration (in m / s²). A 3) on the three axes of the three-dimensional orthogonal coordinate system, forming space. On the x-axis of this curve C2, time is represented in seconds.

[0072]

[0059] Alternatively, the value of the jerk is the sum of the derivatives of the acceleration on one or two axes only of the three axes of the three-dimensional orthogonal frame, forming the space.

[0073]

[0060] It is recalled that in mechanics, a jerk, also a jerk or a jolt, is a sudden variation of the acceleration vector without the notion of a shock, like a driver giving a jolt of the accelerator, a jolt of the brake or a turn of the steering wheel.

[0074]

[0061] System 1 is configured to detect a correlation:

[0075] - between several peaks PR1, PR2, PR3... on the curve C1 obtained by the radar 10 and several peaks PA1, PA2, PA3... on the curve C2 obtained by the accelerometer 40.

[0076]

[0062] These peaks PR1, PR2, PR3... and PA1, PA2, PA3... of the curves C1 and C2 of the radar 10 and the accelerometer 40 are then the characteristics to be compared in the sense of the invention.

[0077]

[0063] For example, a correlation is determined between the peaks PR1, PR2, PR3... on the power curve C1 of the radar 10 and the "jerk" peaks PA1, PA2, PA3... on the accelerometer curve C2 40.

[0078]

[0064] Algorithms that can be used to identify peaks may include, for example (without being limiting): - a simple threshold algorithm that uses a fixed threshold above which a value is considered a peak;

[0079] - an adaptive threshold detection algorithm that dynamically adjusts the threshold according to the variation of the signal, which makes it possible to detect peaks under varying conditions; a derivative detection algorithm determining a zero derivative at a point on the curve, said point being followed by an adjacent point, in the direction of the x-axis, having a negative derivative;

[0080] - a derivative detection algorithm that relies on the fact that peaks often correspond to rapid changes in signal value.

[0081] - a derivative detection algorithm, the algorithm calculates the derivative of the signal and identifies the points where this derivative exceeds a certain threshold.

[0082]

[0065] The determination of the correlation is carried out after filtering the radar 10 data and / or filtering the accelerometer 40 data, for example to retain only the points above a predetermined filtering threshold (for example a filtering threshold equal to 60%, 70%, or even 80%).

[0083]

[0066] System 1, via its computer 20, compares, for example, the amplitude of peaks PR1, PR2, PR3... on curve C1 and the amplitude of peaks PA1, PA2, PA3... on curve C2. The comparison can be performed on a relative amplitude basis because these peaks PR1, PR2, PR3...; PA1, PA2, PA3... of curves C1 and C2 are not measured with the same unit. For example, to perform a relative amplitude comparison, the values ​​of curves C1 and C2 are normalized with respect to their respective maximums. The normalization is performed over the entire time window. This allows for the correlation of entities with different units.

[0084]

[0067] It is still possible that system 1 determines the time interval int1 between two successive peaks among the peaks PR1, PR2, PR3... on the curve C1 and the time interval int2 between two successive peaks among the peaks PA1, PA2, PA3... on the curve C2. If the time interval int1 and the time interval int2 are correlated, then a correlation is concluded and system 1 is configured to ignore the RD result on the presence of life in the vehicle obtained by processing the radar data.

[0085]

[0068] Alternatively, the system 1 is configured to detect a correlation between the shape of curve C1 of the radar data 10 and the shape of curve C2 of the accelerometer data 40. The shape of the curve can, for example, be defined by the slope (rising or falling) of the curve along a predetermined interval. This slope is, for example, given by a derivative calculated on the points of the portion of the curve over this interval. For example, a shape correlation can be concluded if the slope evolves similarly between curves C1 and C2 over a given interval. In another example, the time over which curves C1 and C2 rise or fall can be considered, preferably on a smoothed signal of curves C1 and C2.

[0086]

[0069] Other ways of establishing the correlation can, of course, be considered. For example, one could cite the growth rate or whether the derivative is positive or negative at a point. Another example could be the time span on the two curves C1, C2 to observe a change in the sign of the derivative, on a smoothed signal of the curves C1, C2.

[0087]

[0070] For example, it is possible for system 1 to count the number of peaks PR1, PR2, PR3... on curve C1 and the number of peaks PA1, PA2, PA3... on curve C2, and conclude, from a comparison of the respective number of peaks, whether there is a correlation or not.

[0088]

[0071] Generally speaking, characteristics are chosen which show a good correlation with each other when the radar 10 undergoes its own movements, and the accelerometer 40 measures these movements.

[0089]

[0072] The determination of the correlation is carried out over a time window, or alternatively, at specific points.

[0090]

[0073] The correlation algorithm is configured to use at least one similarity threshold so that above a similarity threshold, a correlation is considered present, and below this similarity threshold, an absence of correlation is considered.

[0091]

[0074] For example, the similarity threshold may be a fixed threshold that allows a comparison to this similarity threshold, a difference of two values ​​of the CR characteristics of the radar 10 and the CA characteristics of the accelerometer 40, or a percentage of one of the values ​​relative to the other value.

[0092]

[0075] For example, the percentage of one of the values ​​relative to the other value is greater than or equal to 70%, preferably greater than or equal to 80%, more preferably greater than or equal to 90%.

[0093]

[0076] According to one aspect of the invention, the accelerometer 40 is configured to measure an acceleration along the three axes (X, Y, Z) of the three-dimensional orthogonal frame, forming space.

[0094]

[0077] In summary, the invention enables the following steps:

[0095] - to process, using the ALP presence detection algorithm, radar data 10 to deliver a result on the presence of life in the vehicle,

[0096] - to process data from accelerometer 40 to deliver a result on a vehicle movement, - to process, using the ALC correlation algorithm, data from radar 10 and data from accelerometer 40 to establish the presence or absence of correlation between at least one CR characteristic from radar 10 data and at least one CA characteristic from accelerometer 40 data, the presence of correlation being representative of a state of inherent movements of radar 10,

[0097] - If a correlation is present, ignore the result regarding the presence of life in the vehicle obtained by processing the radar data. 10

Claims

DEMANDS

1. System (1) for detecting the presence of life inside a motor vehicle (V) by means of a radar (10) and an accelerometer (40), the system being configured to implement the following steps: - to process, in particular using a presence detection algorithm (ALP), radar data (10) to deliver a result on the presence of life in the vehicle, - process accelerometer data (40) to deliver a result on vehicle movement, - to process, in particular by a correlation algorithm (ALC), radar data (10) and accelerometer data (40) to establish the presence or absence of correlation between at least one characteristic (CR) from the radar data (10) and at least one characteristic (CA) from the accelerometer data (40), the presence of correlation being representative of a state of proper motion of the radar (10), - in case of correlation, ignore the result on the presence of life in the vehicle obtained by processing the radar data (10).

2. System according to the preceding claim, wherein, in the absence of correlation, the system (1) is configured to confirm the result on the presence of life in the vehicle obtained by processing the radar data (10).

3. System according to any one of the preceding claims, wherein the correlation algorithm is configured to detect, over a given time window or on a spot basis, a similarity between the signals captured by the radar (10) and the signals captured by the accelerometer (40).

4. System according to any one of the preceding claims, wherein said at least one feature from the radar data (10) to establish the correlation is measured as a function of time, for example so as to define a curve.

5. System according to the preceding claim, wherein said at least one feature from the radar data (10) to establish the correlation is at least one feature of a maximum power curve detected by the radar (10) as a function of time.

6. System according to claim 4, wherein said at least one characteristic derived from accelerometer data (40) to establish the correlation is at least one characteristic of a jerk curve as a function of time.

7. A system according to any one of claims 4 to 6, wherein the system (1) is configured to detect a correlation: - between one or more peaks on the respective curves obtained by the radar (10) and by the accelerometer (40).

8. A system according to any one of claims 4 to 6, wherein the system (1) is configured to detect a correlation: - between the amplitude of the peaks on the respective curves obtained by the radar (10) and by the accelerometer (40).

9. A system according to any one of claims 4 to 6, wherein the system (1) is configured to detect a correlation: - between the time interval between two successive peaks on the respective curves obtained by the radar (10) and by the accelerometer (40).

10. A system according to any one of claims 4 to 6, wherein the system (1) is configured to detect a correlation: - between the shape of the radar data curve (10) and the shape of the accelerometer data curve (40).

11. A system according to any one of the preceding claims, wherein the correlation algorithm is configured to use at least one similarity threshold such that above a similarity threshold, a correlation is deemed to be present, and below this similarity threshold, an absence of correlation is deemed to be present, in particular the similarity threshold being a fixed threshold which allows a comparison to this similarity threshold, a difference of two values ​​of the characteristics of the radar (10) and the accelerometer (40), or a percentage of one of the values ​​relative to the other value.

12. A method for detecting the presence of life inside a motor vehicle (V) using a radar (10) and an accelerometer (40), the method comprising the following steps: - to process, in particular using a presence detection algorithm, radar data (10) to deliver a result on the presence of life in the vehicle, - process accelerometer data (40) to deliver a result on vehicle movement, - to process, in particular by a correlation algorithm, radar data (10) and accelerometer data (40) to establish the presence or absence of correlation between at least one characteristic from the radar data (10) and at least one characteristic from the accelerometer data (40), the presence of correlation being representative of a state of inherent motion of the radar (10), - in case of correlation, ignore the result on the presence of life in the vehicle obtained by processing the radar data (10)

Images