DE102022119526A1 - METHOD AND SYSTEM FOR DETECTING THE PRESENCE OF LIFE IN A VEHICLE - Google Patents

Abstract

The invention relates to a method for detecting the presence of life in a vehicle. The method comprises operating at least one radar sensor device (4) mounted in the vehicle for monitoring a passenger compartment (1) of the vehicle, wherein the operation involves sending out a radar signal in the direction of the passenger compartment of the vehicle and receiving a part of the emitted radar signal that is from a current scenario during a current event in the passenger compartment (1) is reflected. The received radar signal is preprocessed to obtain at least one parameter characterizing the current scenario during the current event. A weight assigned to the at least one parameter is then obtained, the weight being based on a statistical profile for the at least one parameter that indicates a probability of the presence of life in the current scenario during the current event. Based on the weight obtained, the presence of life in the current scenario is determined.

Description

The present invention relates to the field of in-vehicle sensor systems. In particular, the invention relates to a method and a system for detecting the presence of life in a vehicle by at least one radar sensor device.

Modern vehicles, such as cars, are often equipped with a variety of comfort, assistance and safety features and may include interior monitoring systems. Such systems can be designed to detect whether a person is in a seat or not. The prior art has proposed using radar technology for occupant detection systems. This technology can also be used to detect whether a person, particularly a child or baby, is accidentally left in the vehicle, which can be critical to saving lives, such as: B. if the vehicle heats up in the sun. In general, radar sensor technology can be used to detect the presence of life in a vehicle, which may include, for example, pets.

The system may issue some type of signal or alarm message when the presence of life is detected after the vehicle is exited and locked to notify a user, particularly the driver of the vehicle, of a child or pet left in the vehicle. This alarm typically sounds within seconds of the doors closing and locking. For example, the signal or alarm message may be a loud honking of the vehicle horn, or a signal may be sent to a driver's smartphone that attracts the driver's attention. Due to the high level of urgency, the driver must be alerted within seconds or at most a few minutes of leaving the parked vehicle.

To work reliably and safely, such technology would typically not tolerate false negatives, i.e. H. Scenarios in which actual life is incorrectly not detected and no alarm is issued. In particular, a corresponding algorithm must be sensitive enough, e.g. For example, a sleeping baby, especially in an infant car seat with blankets covering the baby, a sleeping infant or small pet, or a child or adult who is almost silent (e.g. because of episodes of respiratory failure occurring). can occur in sleep apnea) or is immobile. However, this can potentially lead to a high false positive rate if the system is oversensitive. While this is more acceptable than false negatives, too many false positives can be annoying for a user and can even cause a user to ignore warnings. False alarms can be caused by other moving objects within the vehicle, particularly when the vehicle is parked and is forced to shake, such as when the vehicle is parked. B. due to strong wind or a truck passing nearby. Such a movable object can be, for example, a filled water bottle or a piece of clothing on a clothes hanger.

An object of the present invention is to provide an improved approach to detecting the presence of life in a vehicle. In particular, it is desirable to improve the accuracy and reliability of detecting the presence of life in a vehicle and, in particular, to reduce the number of false alarms.

A solution to this problem is provided by the doctrine of independent claims. Various preferred embodiments of the present invention are provided by the teachings of the dependent claims.

A first aspect of the invention is directed to a particularly computer-implemented method for detecting the presence of life in a vehicle. The method includes operating at least one radar sensor device mounted in the vehicle to monitor a passenger compartment of the vehicle, the operating comprising transmitting a radar signal toward the passenger compartment of the vehicle and receiving a portion of the transmitted radar signal that reflects a current scenario during a current event in the passenger compartment is included. The received radar signal is preprocessed to obtain at least one parameter characterizing the current scenario during the current event. A weight assigned to the at least one parameter is obtained, the weight being based on a statistical profile for the at least one parameter that indicates a probability of the presence of life in the current scenario during the current event. Based on the weight obtained, the presence of life in the current scenario is determined. If the presence of life is determined, a warning signal can be issued. The presence of life can be detected if the weight is above a predetermined threshold.

Accordingly, the method can be viewed as an improved method for detecting the presence of life (Life Presence Detection - LPD) in a vehicle. The method is particularly advantageous for reducing the number of false alarms. By applying an empirical approach, logic can be implemented that uses weights in sophisticated ways to evaluate the radar system signal.

In other words, the use of statistical profiles instead of heuristic methods simplifies the method of the present invention. There is no need for complicated machine learning that requires greater computational effort and expensive hardware. This leads to a reliable LPD function, which on the one hand can detect the presence of life with high accuracy, while at the same time avoiding or at least reducing false alarms. This can significantly improve safety, especially for children, babies or pets unintentionally left in the vehicle. The statistical profiles can also be used to classify targets, e.g. B. child, adult, pets, water bottles and other common inanimate objects in the vehicle can be used to further improve the process.

The term “detection of the presence of life” as used herein refers specifically to the detection of a living human or a living animal and can be particularly contrasted with the detection of moving or motionless objects. Current technology can be configured to detect even subtle changes within a vehicle to detect life, e.g. B. Movements of a passenger's chest caused by breathing, which may include the capture of breathing patterns. This can apply to adult passengers as well as children or babies and even pets. Thus, the term “presence of life detection” (LPD) used here may specifically include “child presence detection” (CPD). The LPD function can typically be activated after turning off the vehicle's ignition and closing the doors. The term “LPD function” therefore means an activated mode of detecting the presence of life. This function or mode is typically not active, e.g. B. while driving. The LPD function can be activated for a certain period of time after the vehicle is parked (e.g. a few minutes). It may be possible to turn off the LPD function once the presence of life has been detected.

Terms including “radar,” “radar sensor,” “radar sensor device,” “radar sensor system,” or the like as used herein specifically refer to the general understanding of a radar or radar system. More specifically, the “radar” can be a single radar solution or a multiple radar solution in the passenger compartment of the vehicle. Hereafter, the nomenclature “radar” or “radar system” may include single radar or multiple radar solutions. A solution with a radar can be a 3D or a 4D radar. 3D refers to three parameters, namely range, speed, azimuth or elevation angle, while 4D refers to four parameters, namely range, speed, azimuth and elevation angle. A 3D radar is typically used to illuminate only one row of the seats, while a 4D radar can illuminate multiple rows depending on its field of view. One unit of this type of radar may be enough to issue a decision of the LPD function. A multiple radar solution can be composed of several 2D radars. Any 2D radar can only estimate range and speed. They tend to be smaller, less complex and less expensive than a 3D or 4D radar. Several such 2D radars can be placed around the vehicle's cabin to avoid blind spots. There are at least two 2D radars if the locations of moving objects in the car are to be estimated using trilateration techniques. The trilateration process is carried out in a central processing unit. In such a multi-static setup, the decision of the LPD function will most likely come from the calculation in the central processing unit. At the beginning of the process, the radar sends signals to its surroundings. The reflected radar signals are then processed in a pre-processing unit.

The term “scenario,” as used herein, specifically refers to a configuration of a passenger compartment of a vehicle that includes objects in the passenger compartment. Such objects can be living or non-living targets. For example, a current scenario could be a child sitting in a specific seat of the vehicle.

The term “event,” as used herein, specifically refers to actions in a particular scenario that may change over time. Such an event could e.g. E.g. shaking the vehicle, opening or closing doors, etc. In particular, an event does not change the scenario.

The term “vehicle,” as used herein, specifically refers to a car, including any type of motor vehicle, hybrid electric vehicle and battery electric vehicle, as well as other vehicles such as trucks, vans or buses.

Where applicable, the terms "first", "second", "third" and the like are used in the specification and claims to distinguish between similar elements and not necessarily to describe a sequential or temporal order. It is to be understood that the terms so used are interchangeable under appropriate circumstances and that the embodiments of the invention described herein may be practiced otherwise than as described or illustrated herein.

When used in the present description and claims, the term “comprising” or “including” does not exclude other elements or steps. When an indefinite or definite article is used when referring to a singular noun, e.g. E.g. “a” or “an”, “der/die/das”, this includes a plural of this noun unless expressly stated otherwise.

Furthermore, unless expressly stated otherwise, “or” refers to an inclusive or and not an exclusive or. For example, a condition A or B is satisfied by one of the following conditions: A is correct (or present) and B is false (or absent), A is false (or absent) and B is correct (or present) and both A and B are correct (or present).

Preferred embodiments of the method are described below, which can be combined with one another or with other aspects of the present invention as desired, unless such a combination is expressly excluded or is not technically possible.

In some embodiments, the statistical profile includes a table for the at least one parameter for the current scenario during various events. Various events can influence the measurement result of the radar sensor device. The statistical profile thus takes into account different events for a specific scenario.

In some related embodiments, a weight is assigned to the at least one parameter for each of the various events. In this way, the method can reliably determine the presence of life in different events but in the same scenario.

In some embodiments, the statistical profile is based on measurements obtained by operating at least one radar sensor device in a reference scenario with a single target in the passenger compartment of the vehicle during the various events. Reference scenarios with only a single objective can represent simple cases of scenarios, e.g. B. in which only one child sits in the car with no other moving objects nearby.

In some embodiments, the statistical profile is based on measurements obtained by operating at least one radar sensor device in a multi-target reference scenario in the passenger compartment of the vehicle during the various events. Reference scenarios with multiple objectives, e.g. B. a child on a seat and a bottle of water on the floor can provide more complex scenarios that can further improve the results of the LPD procedure. Statistical profiles of single target scenarios and multiple target scenarios can be combined to further improve accuracy.

In some embodiments, a plurality of parameters that characterize the current scenario during the current event are obtained, with weight assigned to a combination of at least some of the plurality of parameters. The presence of more than one parameter along with combined weights can further improve the accuracy of the method since a particular scenario may have a particular combination of characterizing parameters that is different from the combination of parameters for another scenario.

In some embodiments, the at least one parameter includes a radar parameter that includes at least one of a Doppler value, a signal-to-noise ratio, a target range, a target location, and a target frequency movement. It is understood that any common radar parameter can be used.

In some embodiments, the statistical profile further includes a table for an acceleration sensor output parameter for the current scenario during various events. It may be preferred to use data from an acceleration sensor since such sensor data can provide reliable results as to whether the vehicle is being shaken or not.

In some embodiments, the method further includes comparing the obtained weighting with historical data of vibration events of the vehicle, wherein the presence of life is determined based on a result of the comparison. Matching the weighted logic with historical data can improve the accuracy of the method.

In some embodiments, the method further includes comparing the obtained weighting with historical data of actual detection of the presence of life in the vehicle, wherein the presence of life is determined based on a result of the comparison. Matching the weighted logic with historical data can improve the accuracy of the method.

In some embodiments, the method further includes determining a probability of the presence of life in the current scenario based on the weight obtained. In other words, instead of outputting a yes/no decision, a continuous probability from 0 to 1 can be output. This may be preferable for a user to let them ultimately decide whether a child was left behind or not based on probability. It may also be considered to issue different warning levels based on the determined probability.

In some embodiments, the statistical profile is further dependent on a vehicle type. This can further improve the accuracy of the method as the radar results may depend on the vehicle type, e.g. B. Size and shape of the passenger compartment, number and position of seats, etc. In addition, the use of vehicle type-dependent statistical profiles can enable easy adaptation to vehicles with different centers of gravity or damping factors.

A second aspect of the present invention is directed to a data processing system configured to perform the method of the first aspect. The data processing system could be specifically configured by one or more computer programs to perform the method of the first aspect. Additionally or alternatively, the configuration can be implemented in whole or in part using appropriate hardware.

A third aspect of the present invention is directed to a radar sensor system comprising a data processing system according to the second aspect and at least one radar sensor device configured to be mounted in a vehicle to monitor a passenger compartment of the vehicle. It is understood that the radar sensor device may specifically include at least one radar transmitter device for emitting radar waves toward the passenger compartment of the vehicle and at least one radar receiving device for receiving reflected radar waves. The radar sensor device may be configured to emit radar waves toward areas in the passenger compartment of the vehicle where potential passengers can be expected based on positions of seats within the passenger compartment of the vehicle. Of course, it is also conceivable to monitor the entire volume of the vehicle's passenger compartment, e.g. B. to be able to detect pets on the floor.

A fourth aspect of the present invention is directed to a computer program or a computer program product comprising instructions that, when executed on a data processing system according to the second aspect of the invention, cause the system to carry out the method according to the first aspect of the invention.

The computer program (product) can in particular be implemented in the form of a data carrier on which one or more programs for carrying out the method are stored. This is preferably a data carrier, such as a CD, a DVD or another optical medium, or a flash memory module. This can be advantageous if the computer program product is to be traded as an individual product regardless of the processor platform on which the one or more programs are to be executed. In another implementation, the computer program product is provided as a file on a data processing unit, in particular on a server, and can be used via a data connection, e.g. B. the Internet or a dedicated data connection, such as a private or local network.

The system according to the second aspect may accordingly have a program memory in which the computer program is stored. Alternatively, the system can also be set up to access an externally available computer program, for example on one or more servers or other data processing units, via a communication connection, in particular in order to exchange data with it that were used in the course of executing the computer program or the outputs of the computer program.

The explanations, embodiments and advantages described above in connection with the method of the first aspect apply equally to the other aspects of the invention.

• 1 schematisch eine Reihe eines Fahrgastraums eines Fahrzeugs mit einem besetzten Sitz veranschaulicht;
• 2 schematisch zwei Reihen eines Fahrgastraums eines Fahrzeugs mit einem besetzten Sitz und einem zusätzlichen Gegenstand veranschaulicht;
• 3 schematisch Tabellen mit statistischen Profilmustern für verschiedene Szenarien für verschiedene Ereignisse veranschaulicht;
• 4 schematisch eine beispielhafte Ausführungsform eines Verfahrens zum Erstellen von Gewichtungen für Profile während verschiedener Ereignisse veranschaulicht; und
• 5 schematisch eine beispielhafte Ausführungsform eines Verfahrens zum Erfassen des Vorhandenseins von Leben veranschaulicht.

Further advantages, features and applications of the present invention are provided in the following detailed description and accompanying drawings, in which:

• 1 schematically illustrates a row of a passenger compartment of a vehicle with an occupied seat;
• 2 schematically illustrates two rows of a passenger compartment of a vehicle with an occupied seat and an additional item;
• 3 schematically illustrates tables with statistical profile patterns for different scenarios for different events;
• 4 schematically illustrates an exemplary embodiment of a method for creating weights for profiles during various events; and
• 5 schematically illustrates an exemplary embodiment of a method for detecting the presence of life.

1 schematically illustrates a passenger compartment 1 of a vehicle with an example of a single target profile measurement arrangement of a child 2 sitting on a seat 23. In this example, the child 2 sits on the right rear seat 23 of the three seats 21, 22, 23 of the rear seat of the vehicle. The measurement is carried out by a radar sensor device 4 aimed at the back seat of the vehicle. An acceleration sensor 5 may be provided to further verify a vehicle shaking event, as explained in more detail below. The acceleration sensor chip 5 can be located on the radar sensor device 4.

2 analogously schematically illustrates another profile measuring arrangement. In this case, a multi-target profile of a water bottle 3 on the floor of the passenger compartment 1 of the vehicle with a child 2 in the right rear seat 23 is determined. 2 For the sake of completeness, also shows the driver's seat 11 and the passenger seat 12.

The measured single target profile and the measured multiple target profile result in different parameter patterns as set out below. First, general terms are explained in more detail for a better understanding of the invention. In the following description, a distinction can be made between “normal cases” and “corner cases” that may arise during the operation of the life presence detection (LPD) function.

The term “normal case,” as used herein, specifically refers to cases where a vehicle is parked in a normal location, such as a parking garage or a designated open parking space. It can be assumed that there will be little to no external disturbances that cause the car to shake, such as heavy trucks driving past. Typically, the LPD function turns on as soon as the parked vehicle ignition is turned off and the doors are closed. In such a case, which probably occurs most of the time, the LPD function can detect the presence of a child, pet or person almost 100% correctly.

The term “borderline case”, as used here, refers in particular to cases that are not “normal cases”. Such a “borderline case” or scenario should subsequently lead to a “vibration event” of the parked vehicle. This can e.g. This may be the case, for example, if the vehicle is parked on a busy street. On this street, there may be opportunities for external disturbances that may cause the parked vehicle to shake. This can happen when a car or truck drives past the parked vehicle. The difference in air pressure causes air to be displaced, which then shakes the parked vehicle.

A “borderline” or “shock event” may have consequences for LPD function that are addressed by the present invention. For example, if a vibration event occurs on a parked vehicle, a radar sensor system may tend to misidentify the signals it detects.

A false positive alarm can occur if there are loose parts or objects in the vehicle that are in an oscillating motion, such as oscillating and vibrating. Examples of the most common objects that exhibit such swinging movements include water in a bottle (glass bottle, milk bottle, sippy cup, etc.), items on hangers hanging from the handle above the door, and toys. The oscillating movement can resemble a human's breathing pattern. Therefore, the LPD function on a radar may not be able to distinguish between a real human and an inanimate object. If the oscillating motion happens to be within the range of a human's breathing rate, it will show up as detected life.

A false negative result can also occur in almost the same way. If the shaking event causes the entire car, as well as the inanimate objects within the car, to shake with an intensity greater than a human's breathing rate, the LPD may derive mixed signals that may lead to a determination that no life has been detected. While it is true that the oscillation of water in a bottle or a coat hanging on a coat hanger in the car will decay and then stop over time, there is a possibility that it may not be fast enough (within the time limit) for such a time-critical application to activate the alarm message). The present invention particularly focuses on mitigating false alarms in the corner cases with an LPD algorithm that uses radar.

Subsequently, the proposed LPD method can implement a logic that can be referred to as “weighted decision logic”. As explained in the following examples, the most common outputs (parameters) from a generic radar can be used together with the LPD algorithm with weighted decision logic to mitigate false alarms. It has been found that with appropriate parameter settings, false alarms can be reduced up to 99.99% of the time under realistic circumstances. Complicated machine learning can be avoided to reduce costs in hardware and R&D efforts. As explained above, an optional acceleration sensor chip 5 can also be used. With an acceleration sensor 5 in the radar sensor device 4, a further source of information about the vibration events on the parked vehicle is available. This information can be used to refine the decision logic of the LPD to achieve better decision accuracy and make faster decisions.

The proposed weighted decision logic LPD algorithm can be used with any radar using any type of waveform as long as certain common parameters appear at the output of the radar's preprocessing unit. This is explained below. It is also imperative that the radar's parameters are configured to match the characteristics of its environment. For example, the range and speed resolution must be small enough to capture detections in a small space of the vehicle's cabin. In addition, enough antenna elements should be used to achieve a sufficiently fine angular resolution. In the exemplary embodiment described herein, a radar operating at 60 GHz with 3Tx-4Rx was used to validate the LPD algorithm.

With reference to 3 Example tables (a), (b) and (c) of profile parameters with parameters 1...N are provided for various scenarios in the passenger compartment 1 of a vehicle during various events. The parameters refer to common radar parameters (outputs) and may include at least one of the parameters A, B, C, D and E listed below.

Common outputs of the preprocessing unit may be (A) power or signal-to-noise ratio (SNR) of the multiple targets. This can include a bin for power per range/Doppler/angle or average power. Other parameters may include (B) range (distance from radar to targets) or (C) Doppler or velocity (from targets relative to radar). Apart from this, (D) locations of multiple targets can be considered, e.g. B. in terms of azimuth and/or elevation angle, spherical coordinates with [R,θ,φ]-axis, where R = radius, θ,φ are the azimuth and elevation angle, or Cartesian coordinates with [x,y,z]- Axis. Furthermore, (E) frequency movements of the targets can be taken into account. This can be divided into normal respiratory rate - the range includes respiratory rates for babies, children, adults and common pets; high or low frequency movements - most likely inanimate objects such as toys or bottled water; random movements - where no regular frequency pattern can be identified; Harmonics - frequencies induced by the fundamental frequency or true frequency (a harmonic is always n times the fundamental frequency, where n = 2, 3, 4, ...). A harmonic frequency is most likely the result of mechanical vibrations rather than normal breathing. The above-mentioned parameters A, B, C, D, E can be used for each data bin, i.e. H. Am available for per range/Doppler/angle.

It is common to use data buffers to accumulate data over time. The data can be the raw or processed data of various processing stages. There can be at least two data buffers that store the data for different periods of time. Shorter buffers store less data and are used to produce faster results. These results may be erroneous due to the short observation time. Longer buffers store more data over a longer period of time and are used to produce more accurate results. The reason for this is that observation over a longer period of time results in a more reliable picture of the surroundings. The data buffer sizes can be worked out based on the radar waveforms and timings used. This is to ensure that a sufficiently long observation time is maintained for the shorter buffers to obtain meaningful answers for frequency and power. The size of the longer buffers should not exceed the limit at which the state of objects and people in the cabin could have changed significantly. The above-mentioned parameters A, B, C, D, E may be available for each data bin, ie bin per range/Doppler/angle and per buffer.

In the 3a , 3b and 3c The profile patterns shown for each event are composed of exemplary parameter combinations. The parameters referred to as Param 1, 2, 3, 4, ..., N can be the common radar parameters specified above under A, B, C, D, E, as well as any external parameter such as accelerometer, vehicle host data, camera data, etc. or these include. The example profile patterns show different events along the timeline 0...19, including door opening and closing (time 1-2), shock event (time 6-12), stationary (no shock) (time 16-19). After a start at time 0, some transition times from one state to another (times 3-5 and times 13-15) occur. In the respective time columns, an “x” indicates in the respective parameter row whether the radar sensor device 4 has an output signal in this regard.

3a shows a statistical profile related to a child present in a vehicle, as in 1 illustrated. 3b shows a profile of just one water bottle (cf. water bottle as in 2 shown). 3c shows a profile of the water bottle and child who are in 2 are shown. The profile seen here is usually a mix of the other two profiles mentioned above. This is used as confirmation and reconciliation of the weighted probabilities.

An example method 100 for building weights will now be described with reference to 4 described. This methodology uses the obtained tread patterns from mentioned above 3 . The start and end are indicated as S11 and S16, respectively.

First, measurement campaigns (S12) are conducted to obtain statistical profiles of individual targets such as adult, child, pet, bottled water, toys and other common inanimate objects in the car as in 1 illustrated. These measurements are made for various events, including both shock and non-shock events on the vehicle. Each profile consists of a “pattern” (see 3a , 3b) , which can be a combination of different parameters of A, B, C, D, E from each data buffer.

Next, statistical profiles of common multiple targets are determined (S13). Examples include a child with a milk bottle/toy/pet, clothes hangers on the handrail, water bottles on the floor, or unpadded rooms (see 2 ). In the same way, the measurements are made for various events, including both shock and non-shock events on the vehicle (cf. 3c ).

Then tables like those in 3a , 3b and 3c shown to represent statistical probabilities (S14). They are then weighted based on their statistical probability of appearing in the multiple profiles (S15). In other words, weights are assigned to the parameters or combinations of parameters for the various events.

If the radar has outputs from each data buffer in addition to the parameters A, B, C, D, E, these can also be taken into account to build the profiles. If additional redundancy information is available in addition to the parameters and acceleration sensor data mentioned above, this can be taken into account when building the profiles. Examples include camera data, intelligent cabin access sensors that detect motion, vehicle host data.

For example, Param 3 and 4 are made out 3a and 3b shown in both single target profiles during a concussive event. A review of the multi-target profile in 3c confirms this mixture of Param 3 and Param 4. The weighted logic in this case can look like this in the algorithm: if (Param 3 && Param 4), then life_detection = 0.5. Or in words, if Param 3 and Param 4 occur together, the probability of detecting life due to human being is 50%.

When the parameters or combination of parameters have been identified with a statistical probability assigned to all events based on the individual target profiles, as explained using the example above, the methodology ends (S16). The next stage may be the top-level weighted decision logic algorithm, described below with reference to 5 is described.

With reference to 5 will a method 200 for detecting the presence of life using an exemplary implementation of an LPD algorithm with weighted decision logic starting at S21 and ending at S28 is explained. In this top-level algorithm, the weighted logic is compared with historical data to make the final decision whether to detect life or not. First, radar preprocessing with output of Param 1, 2, 3, ..., N is carried out (S22). This can be understood as executing the LPD function after the vehicle doors have been closed. The following steps are taken to make a decision whether there is life in the vehicle or not. Based on the assigned weights as described above, a decision in the form of a probability between 0 and 1 is output (S23). The probability of can e.g. B. be 0.5.

The historical data on the statistical probability of a life being recorded is also accumulated over time. When using an accelerometer, historical data is also recorded to capture vibration events. In a step 24 it can then be determined whether the course of a shock event is above a certain threshold and, if so, a final decision with a probability between 0 and 1 can be issued (S26). For example, in this case it might be more likely that there is no life present, but that this was just a concussive event. However, continuing with step S25, it can then be determined whether the course of detected life is above a certain threshold value. If yes, a final decision can be issued with a probability between 0 and 1 (S27). For example, an alarm signal can be issued that the presence of life has been detected. It is beneficial if the final logic depends on the historical data as the confidence level will be higher.

Using a continuous probability from 0 to 1 instead of just "0" or "1", especially in the first few seconds when the reliability of the output is low (due to lack of historical data), can be beneficial because it allows you to assign different levels of warning based on continuous probability. This could be used to issue different alarm messages (e.g. loud horn or just a light signal) to the user.

The above methodology can be used to classify as child, adult, pet, water bottle and other common inanimate objects in the car. A simple adaptation to vehicles with different centers of gravity or damping factors may be possible. While it is understood that machine learning can be used in this method, there is no absolute need for complicated machine learning that requires greater computational effort and expensive hardware.

While at least one exemplary embodiment of the present invention has been described above, it should be noted that a large number of variations thereof exist. Furthermore, it is to be understood that the exemplary embodiments described merely illustrate non-limiting examples of how the present invention may be implemented and are not intended to limit the scope, application, or configuration of the devices and methods described herein. Rather, the foregoing description provides those skilled in the art with constructions for implementing at least one exemplary embodiment of the invention, it being understood that various changes in the functionality and arrangement of the elements of the exemplary embodiment may be made without departing from the appended claims and their legal rights Equivalents deviate from the defined item.

REFERENCE SYMBOL LIST

1

Passenger compartment of the vehicle

2

child

3

Bottle

4

Radar sensor device

5

Accelerometer

11

Driver's seat

12

Passenger seat

21

left rear seat

22

middle back seat

23

right rear seat

Claims (15)

Method for detecting the presence of life in a vehicle, the method comprising: - operating at least one radar sensor device (4) mounted in the vehicle for monitoring a passenger compartment (1) of the vehicle, the operation emitting a radar signal in the direction of the passenger compartment of the vehicle and receiving a part of the transmitted radar signal which is reflected by a current scenario during a current event in the passenger compartment (1); - Pre-processing the received radar signal to obtain at least one parameter characterizing the current scenario during the current event: - Obtaining a weight assigned to the at least one parameter, the weight being based on a statistical profile for the at least one parameter that represents a probability indicates the presence of life in the current scenario during the current event; and - determining the presence of life in the current scenario based on the weight obtained. Procedure according to Claim 1 , wherein the statistical profile includes a table for the at least one parameter for the current scenario during various events. Procedure according to Claim 2 , whereby a weight is assigned to the at least one parameter for each of the different events. Method according to one of the preceding claims, wherein the statistical profile is based on measurements obtained by operating at least one radar sensor device (4) in a reference scenario with a single target in the passenger compartment (1) of the vehicle during the various events. Method according to one of the preceding claims, wherein the statistical profile is based on measurements obtained by operating at least one radar sensor device (4) in a reference scenario with multiple targets in the passenger compartment (1) of the vehicle during the various events. A method according to any one of the preceding claims, wherein a plurality of parameters are obtained that characterize the current scenario during the current event, the weight being assigned to a combination of at least some of the plurality of parameters. Method according to one of the preceding claims, wherein the at least one parameter comprises a radar parameter comprising at least one of a Doppler value, a signal-to-noise ratio, a target range, a target location and a target frequency movement. A method according to any one of the preceding claims, wherein the statistical profile further includes a table for an output parameter of an acceleration sensor for the current scenario during various events. A method according to any one of the preceding claims, further comprising comparing the obtained weight with historical data of vibration events of the vehicle, wherein the presence of life is determined based on a result of the comparison. A method according to any one of the preceding claims, further comprising comparing the obtained weighting with historical data of actual detection of the presence of life in the vehicle, wherein the presence of life is determined based on a result of the comparison. A method according to any one of the preceding claims, further comprising determining a probability of the presence of life in the current scenario based on the weight obtained. Method according to one of the preceding claims, wherein the statistical profile is further dependent on a vehicle type. Data processing system, the system being configured to implement the method according to one of Claims 1 until 12 to carry out. Radar sensor system, comprising a data processing system Claim 13 and at least one radar sensor device (4) configured to be mounted in a vehicle to monitor a passenger compartment (1) of the vehicle. Computer program or computer program product comprising instructions that follow when executed on a data processing system Claim 14 cause the system to carry out the procedure according to one of the Claims 1 until 13 executes.

Images