EP3327690B1 - Improved fall detection device - Google Patents

Improved fall detection device Download PDF

Info

Publication number
EP3327690B1
EP3327690B1 EP17203320.1A EP17203320A EP3327690B1 EP 3327690 B1 EP3327690 B1 EP 3327690B1 EP 17203320 A EP17203320 A EP 17203320A EP 3327690 B1 EP3327690 B1 EP 3327690B1
Authority
EP
European Patent Office
Prior art keywords
fall
threshold
values
calculating
calculation
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
EP17203320.1A
Other languages
German (de)
French (fr)
Other versions
EP3327690A1 (en
Inventor
Emmanuel RAVON
Didier Filhol
Aurélien GIBIAT
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Telecom Design SA
Original Assignee
Telecom Design SA
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Telecom Design SA filed Critical Telecom Design SA
Publication of EP3327690A1 publication Critical patent/EP3327690A1/en
Application granted granted Critical
Publication of EP3327690B1 publication Critical patent/EP3327690B1/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G08—SIGNALLING
    • G08B—SIGNALLING OR CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
    • G08B21/00—Alarms responsive to a single specified undesired or abnormal condition and not otherwise provided for
    • G08B21/02—Alarms for ensuring the safety of persons
    • G08B21/04—Alarms for ensuring the safety of persons responsive to non-activity, e.g. of elderly persons
    • G08B21/0438—Sensor means for detecting
    • G08B21/0446—Sensor means for detecting worn on the body to detect changes of posture, e.g. a fall, inclination, acceleration, gait
    • G—PHYSICS
    • G08—SIGNALLING
    • G08B—SIGNALLING OR CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
    • G08B21/00—Alarms responsive to a single specified undesired or abnormal condition and not otherwise provided for
    • G08B21/02—Alarms for ensuring the safety of persons
    • G08B21/04—Alarms for ensuring the safety of persons responsive to non-activity, e.g. of elderly persons
    • G08B21/0407—Alarms for ensuring the safety of persons responsive to non-activity, e.g. of elderly persons based on behaviour analysis
    • G08B21/043—Alarms for ensuring the safety of persons responsive to non-activity, e.g. of elderly persons based on behaviour analysis detecting an emergency event, e.g. a fall

Definitions

  • the present invention relates to a method and a device for monitoring a person carrying the device and for detecting falls of this person, in particular for the elderly but also isolated workers, athletes and any person where the situation of falling represents a risk.
  • the detection of a fall must trigger an alarm with a surveillance service or with relatives.
  • Fall detection devices exist and in particular the document EP 1 642 248 A1 discloses a portable device that detects a fall by detecting a difference in height by means of a barometric measurement. Such a measure requires the use of an external reference to eliminate false detections caused by natural variations in pressure, such as those caused by door closures, drafts, particularly those resulting from building ventilation / heating systems. and temperature variations. In addition the power consumption of a permanent barometric device remains significant and much greater than that of an accelerometer.
  • the extraction of the gravity component consists in estimating the gravity vector g by using a sphere of unit radius discretized in several vertices to model the set of possible directions of the unit vector g, and a stochastic calculation to determine the probability carried by each vertex that the vector g is in the direction indicated by this vertex.
  • Other relevant prior art documents are CN 102 610 056 B , US 2013/054180 A1 and CN 102 393 992 A .
  • the present invention relates to an improved fall detection device and for which the detection and recognition of a fall is improved.
  • the method comprises a detection of gravity loss sequences, gravity feedback, loss of gravity so as to filter movements opposed to the general direction of a fall.
  • the method may further comprise a filtering of repetitive movements.
  • the validation or non-validation step may comprise a series of pressure measurements and a pressure variation calculation adapted to detect a lift of the wearer.
  • the method includes a delay before alarm called first timer.
  • the method may include a recovery detection wait for an alarm disable sensor during said first time delay and a second recovery time validation time delay, the alarm being triggered if a support is not detected during the first timeout. delay or if said overlap is of a duration less than the second time delay.
  • the detection of a worn state and an unworn state of the device by a user is performed by measuring the variation of the I / Q signal of a radio component of the device.
  • the invention may in particular provide a modeling of the wearer's steps by the analysis of the small recurrent shocks due to the steps which gives the manner in which it moves, said modeling comprising the modeling of the parameters speed of walking, amplitude of the steps, regularity, imbalance possible gait and having one or more signaling stages of significant variations of said parameters may reflect a lack of balance that may lead to a risk of falling.
  • the method of the invention may include the activation of a vibrator at the end of the fall detection validation step.
  • the invention further relates to a portable device for monitoring and detecting falls of a person adapted to implement the method of the invention and which comprises a three-axis accelerometer for detecting falls, a deactivation sensor of alarm, a pressure sensor, a buzzer, calculation means and alarm means.
  • the device may include a battery-powered sealed housing that can be worn as a medallion, bracelet, or belt.
  • the device can also be miniaturized to be incorporated in a garment or in any item worn.
  • the alarm means may include radio components adapted to transmit the alarms and the operating data of the device either to a mobile phone or to a remote support base or to a LPWAN (Low Power Wide Area Network) network.
  • LPWAN Low Power Wide Area Network
  • the device and the method of the present invention have the primary objective of helping to keep elderly people at home by preventing the risks of falling by step analysis, by detecting falls if they happen anyway and by sending alarms by radio. to a base connected to a monitoring center or to a mobile telephone equipped with a remote monitoring application in the event of a fall or abnormal behavior. They can also be used for the protection of lone workers or outdoor sportsmen and anyone who is in danger of falling.
  • the device of the invention comprises a sealed housing, similar to a watch case for example, powered by a rechargeable battery or battery which can be worn as a medallion, a bracelet or a belt but which can be miniaturized to be incorporated into a garment or in any item worn.
  • the device is equipped with radio components 5 and an antenna 8 for transmitting alarms and operating data of the device either in low energy Bluetooth (Bluetooth Low Energy) to a mobile phone for example, or in radio in the ISM band 868 Mhz for the dialogue with a base of tele-assistance or a LPWAN network (Low Power Wide Area Network) for example.
  • Bluetooth Bluetooth Low Energy
  • ISM band 868 Mhz for the dialogue with a base of tele-assistance
  • LPWAN network Low Power Wide Area Network
  • a remote assistance base In the case of a remote assistance base, the latter is arranged in a fixed position in the building in which the person carrying the device is located so that the device can interact with the base and transmit the detection information.
  • the device can communicate with either a mobile phone or an LPWAN network to transmit alarms.
  • the device comprises in particular a 3-axis accelerometer for detecting falls, a sensitive sensor 6 to enable an alarm to be deactivated when the person wishes, a pressure sensor 4 which will serve to confirm a fall or, on the contrary, to invalidate it and a vibrator 9.
  • the device comprises processing and calculation means in the form of a microprocessor or microcontroller 1 and data storage means / program 2 in RAM and ROM or reprogrammable to store the data and the internal program.
  • the device performs a signal processing provided by the accelerometer and other means of detection and alarm.
  • All the components of the device are conventionally arranged on a printed circuit and connected by electric tracks in the form of one or more buses 7 and the processing and calculation means comprise one or more digital-to-analog converters for processing the data.
  • accelerometer and pressure sensor for example.
  • the device is powered by a button-type battery or a rechargeable battery.
  • Verification of the actual port of the object is done by studying the variation of the I / Q signal given by the radio component 5. Indeed the presence of a human body in the vicinity of the device will modify the I / Q signal by changing the tuning of the antenna 8 of the device which can detect whether the device is worn or not.
  • the fall alarm is signaled to the carrier by the vibrator 9 and the carrier can possibly disable it by putting your hand on the sensor sensor 6 before the alarm is transmitted.
  • the device further comprises two LEDs 11, red LEDs to signal a fall detection and / or the sending of an alarm, a green LED 12 linked to the operation and a push button 13 for manually triggering an alarm.
  • a pseudo-vertical position is set up to reduce false alarms as long as this orientation model is valid.
  • the modeling of the position can be performed since the device maintains a pseudo verticality.
  • a variance calculation on X, Y, Z over a period of the order of one second characterizes the stability of the orientation in the space of the wearer.
  • a model of this orientation is established and updated from the averages of X, Y, Z, retaining only sufficiently stable orientations.
  • the follow-up of this orientation model makes it possible to isolate limited orientation variations, even interspersed with brief rapid variations, which correspond to a controlled activity. This detection makes it possible to cancel a fall, if the orientation of the device respects the established model, or to confirm it if the model is no longer validated.
  • the device will model the wearer's steps by analyzing, using the accelerometer 3 small recurring shocks due to the steps that gives the manner in which it moves and in particular, the walking speed, the amplitude steps, their regularity, the possible imbalance of the march and to announce important variations likely to translate a lack of equilibrium which could lead to a risk of falling.
  • the fall detection itself is performed by analyzing the measurements acquired by the accelerometer 3.
  • the accelerometer transmits a measurement combining the acceleration due to gravity and the acceleration due to the movements of the wearer of the device.
  • the accelerometer data in the 3 axes (X, Y, Z) of the accelerometer are mainly worked in standard 2, ie in the form X 2 + Y 2 + Z 2 .
  • the X, Y and Z axes of the accelerometer are fixed relative to the accelerometer whose position in space is arbitrary.
  • the measurements given by the accelerometer for the 3 axes are a combination of the gravity component and motion components.
  • the fall detection principle of the present invention is schematized in the figure 2 . It consists in analyzing the measurements through three study prisms: the search for the free fall of the device 100 corresponding to a cancellation of the gravity which is symptomatic of a free fall, the analysis of the movements 200 of the person, the fact that the fall of a person is accompanied by movements, and the detection of shocks 300 by detection of the rebound of the accelerometer weight and the variation of the angle of the vector represented by X, Y and Z in the time by means of a computation of the solid angle formed between 2 temporally close samples, this detection being necessary to determine an end of fall.
  • a first series of computations corresponds to the search for the free-fall and comprises a calculation 101 of the norm squared N 2 of the positions X 2 + Y 2 + Z 2 , a storage 102 of n1 values of N 2 .
  • step 205 the free-fall indicator of step 106 is taken into account, which, if it is set, will set the value N 2 ⁇ to a predetermined maximum value.
  • step 206 is performed a comparison of N 2 ⁇ with a threshold S ⁇ .
  • a third series 300 of calculations corresponds to the detection of shocks and comprises a storage 301 of n 4 values of X, Y, Z, the calculation 302 of the standard squared X 2 + Y 2 + Z 2 , the calculation 303 of the variation of the angle V a between the samples and the comparison 304 of this variation V a with a threshold S a .
  • N 2 ⁇ When Va is greater than the threshold Sa, the value of N 2 ⁇ is validated at 207 and the values N 2 , N 2 ⁇ are used in a mathematical function 400 giving a drop relevance value itself compared to a threshold 410 to characterize the fall.
  • the mathematical function which takes into account the value filtered in step 105 of c and the square standard of ⁇ (N 2 ⁇ ) validated (207), integrates in particular a multiplication of these 2 values (y 1 ), the subtraction of a constant to the multiplication result to eliminate minor events (y 2 ).
  • This function makes it possible to detect an exceeding of the threshold followed by a return to a normal state in a determined time characteristic of a fall or at least the end phase of a fall (before even stopping briefly when it is followed by convulsions )
  • the principle of squared standard data analysis makes it necessary to take into account a certain number of phenomena and to filter them. In particular the scaling of noise (amplified by the square), the recovery of signed values when necessary
  • the accelerometer will detect an acceleration which can be interpreted as a return of the gravity component while we are in front of a fall. For this we will isolate the sequences, loss of gravity, return of gravity, loss of gravity.
  • This filtering integrates an autocorrelation of the signals X, Y, Z taken separately on different time scales to detect several characteristic frequency spectra eg in a range of 0.3 to 3 Hz
  • the device acquires the data of the accelerometer in X, Y, Z between 20Hz and 30Hz and more particularly around 25Hz and the accelerometer comprises a low pass filter between 1000Hz and 2000Hz, typically at 1500Hz.
  • the latter is equipped with connection means to the telephone network, radio means compatible with the radio means of the device, processing and calculation means adapted to perform an alarm management and monitoring the device.
  • the device is optimized to have a low power consumption which is possible by limiting the use of a pressure sensor only in the research phase of a recovery of the device to invalidate an alarm.
  • the device can possibly send specific information which can for example of minor alert for the surveillance personnel.
  • the combination of squared standard measurement with shock detection and motion analysis measures substantially improves the filtering of false fall detections compared to prior devices with accelerometer sensors.
  • the invention is not limited to the example shown and in particular the device in a more complex version may include other sensors such as a GPS sensor.
  • a standby mode of inactivity detection for a few seconds completes the system, reducing energy consumption.
  • the sleep and wake-up phase manages all the necessary updates and invalidation of the memorized elements.

Description

Domaine de l'inventionField of the invention

La présente invention concerne un procédé et un dispositif de surveillance d'une personne porteuse du dispositif et de détection de chutes de cette personne, notamment pour les personnes âgées mais aussi les travailleurs isolés, les sportifs et toute personne où la situation de chute représente un risque. Pour ce type de dispositifs, la détection d'une chute doit déclencher une alarme auprès d'un service de surveillance ou auprès des proches.The present invention relates to a method and a device for monitoring a person carrying the device and for detecting falls of this person, in particular for the elderly but also isolated workers, athletes and any person where the situation of falling represents a risk. For this type of device, the detection of a fall must trigger an alarm with a surveillance service or with relatives.

Arrière plan technologiqueTechnological background

Des dispositifs de détection de chutes existent et notamment le document EP 1 642 248 A1 décrit un dispositif portable qui détecte une chute par détection de différence de hauteur au moyen d'une mesure barométrique. Une telle mesure oblige à utiliser une référence externe pour éliminer les fausses détections causées par les variations naturelles de pression comme celles causées par les ouvertures fermetures de portes, les courants d'air en particulier ceux issus des systèmes d'aération/chauffage de bâtiments, et les variations de température. En outre la consommation électrique d'un dispositif barométrique permanent reste importante et bien supérieure à celle d'un accéléromètre.Fall detection devices exist and in particular the document EP 1 642 248 A1 discloses a portable device that detects a fall by detecting a difference in height by means of a barometric measurement. Such a measure requires the use of an external reference to eliminate false detections caused by natural variations in pressure, such as those caused by door closures, drafts, particularly those resulting from building ventilation / heating systems. and temperature variations. In addition the power consumption of a permanent barometric device remains significant and much greater than that of an accelerometer.

D'autres méthodes emploient un accéléromètre comme dans le document EP 2 207 154 A1 qui concerne:

  • un dispositif de détection de chute pour un utilisateur, comprenant un boîtier renfermant des moyens d'alimentation, des moyens de transmission d'un signal d'alerte, un accéléromètre, et des moyens de traitement des mesures fournies par l'accéléromètre comprenant des moyens de calcul et des moyens de mémorisation et,
  • un procédé prévoyant la réalisation par les moyens de traitement d'une chaîne de traitement des mesures acquises par l'accéléromètre qui est composée des étapes suivantes:
    • normalisation des mesures d'accélération en données d'accélération,
    • extraction de la composante gravité desdites données,
    • transformation et agrégation desdites données en indicateurs, tant qualitatifs que quantitatifs, de l'activité de l'utilisateur,
    • classification du comportement de l'utilisateur à l'aide desdits indicateurs,
    • déclenchement d'un signal d'alerte en cas d'observation d'un comportement supposé anormal de l'utilisateur.
Other methods use an accelerometer as in the document EP 2 207 154 A1 Concerning:
  • a fall detection device for a user, comprising a housing enclosing supply means, means for transmitting an alert signal, an accelerometer, and means for processing measurements. provided by the accelerometer comprising calculation means and memory means and,
  • a method providing the processing means with a processing chain of the measurements acquired by the accelerometer which is composed of the following steps:
    • standardization of acceleration measurements in acceleration data,
    • extracting the severity component of said data,
    • transformation and aggregation of said data into indicators, both qualitative and quantitative, of the user's activity,
    • classification of the user's behavior using said indicators,
    • triggering an alert signal in case of observation of a supposedly abnormal behavior of the user.

Selon ce document, l'extraction de la composante gravité consiste à estimer le vecteur gravité g en utilisant une sphère de rayon unitaire discrétisée en plusieurs sommets pour modéliser l'ensemble des directions possibles du vecteur g unitaire, et un calcul stochastique pour déterminer la probabilité portée par chaque sommet que le vecteur g se trouve dans la direction indiquée par ce sommet. D'autres documents de l'état de la technique pertinents sont CN 102 610 056 B , US 2013/054180 A1 et CN 102 393 992 A .According to this document, the extraction of the gravity component consists in estimating the gravity vector g by using a sphere of unit radius discretized in several vertices to model the set of possible directions of the unit vector g, and a stochastic calculation to determine the probability carried by each vertex that the vector g is in the direction indicated by this vertex. Other relevant prior art documents are CN 102 610 056 B , US 2013/054180 A1 and CN 102 393 992 A .

Quelque soit la méthode employée un problème important est l'existence de fausses détections.Whatever the method employed, an important problem is the existence of false detections.

Brève description de l'inventionBrief description of the invention

La présente invention a pour objet un dispositif de détection de chute perfectionné et pour lequel la détection et la reconnaissance d'une chute est améliorée.The present invention relates to an improved fall detection device and for which the detection and recognition of a fall is improved.

Plus particulièrement, l'invention prévoit un procédé de surveillance de l'état d'une personne porteuse d'un dispositif de surveillance et de détection de chutes réalisé par analyse de mesures acquises par un accéléromètre trois axes qui transmet une mesure combinant une accélération due à la gravité et une accélération due aux mouvements de ladite personne qui comprend:

  • un calcul de recherche de chute libre comportant le calcul de la norme au carré N2= X2+Y2+Z2 d'échantillons successifs de données de mesure des axes X, Y et Z de l'accéléromètre, un stockage de n1 valeurs de N2, un calcul d'une valeur moyenne a des n1 valeurs de N2 et une valeur moyenne b des p dernières valeurs de N2, un calcul (104) de la différence c=a-b, un filtrage;
  • la comparaison de valeurs successives de la norme au carré N2 par rapport à un seuil pour positionner un indicateur de chute libre si n2 valeurs successives de N2 sont inférieures au seuil;
  • un calcul d'analyse du mouvement comportant un stockage de n3 échantillons de X, Y, Z, le calcul de la différence Δ entre [des échantillons temporellement rapproché,, le calcul (203) de la norme au carré de ladite différence Δ: N2Δ= ΔX2+ ΔY2+ ΔZ2, un filtrage des valeurs N2Δ, le remplacement des valeurs N2Δ calculées par une valeur prédéterminée si l'indicateur Chute libre est positionné, la comparaison de N2Δ avec un seuil SΔ;
  • un calcul de détection de chocs comportant un stockage de n4 valeurs de X, Y, Z, le calcul de la norme au carré N2=X2+Y2+Z2, le calcul de la variation de l'angle Va entre les échantillons par un calcul d'angle solide adapté aux normes carrées],
  • une étape de comparaison de cette variation Va à un seuil Sa telle que:
    • lorsque Va est supérieur au seuil Sa, la valeur de N2Δ est validée en tant que choc et les valeurs N2, N2Δ sont utilisées dans une fonction mathématique intégrant un seuil et une dérivation de Va afin de détecter un dépassement du seuil suivi d'un retour à un état normal dans un temps déterminé caractéristique d'une chute donnant une valeur de pertinence y de chute elle même comparée à un seuil Sa pour caractériser la chute,
    • lorsque y est inférieur au seuil Sa la chute n'est pas suspectée, lorsque y est supérieur au seuil, la chute est suspectée.
More particularly, the invention provides a method of monitoring the state of a person carrying a monitoring device and detection of falls made by analysis of measurements acquired by a three-axis accelerometer that transmits a measurement combining an acceleration due gravity and acceleration due to movements of said person that includes:
  • a free-fall search calculation comprising the calculation of the squared norm N 2 = X 2 + Y 2 + Z 2 of successive samples of measurement data of the X, Y and Z axes of the accelerometer, a storage of n1 values of N 2 , a calculation of an average value has n 1 values of N 2 and a value mean b of the last p values of N 2 , a calculation (104) of the difference c = ab, a filtering;
  • comparing successive values of the norm squared N 2 with respect to a threshold for positioning a free fall indicator if n2 successive values of N 2 are below the threshold;
  • a motion analysis calculation comprising a storage of n3 samples of X, Y, Z, the calculation of the difference Δ between [temporally close samples, the calculation (203) of the norm squared of said difference Δ: N 2 Δ = ΔX 2 + ΔY 2 + ΔZ 2 , a filtering of the values N 2 Δ, the replacement of the values N 2 Δ calculated by a predetermined value if the indicator Free fall is set, the comparison of N 2 Δ with a threshold S Δ ;
  • a shock detection calculation comprising a storage of n4 values of X, Y, Z, the calculation of the norm squared N 2 = X 2 + Y 2 + Z 2 , the calculation of the variation of the angle V a between the samples by a solid angle calculation adapted to square standards],
  • a step of comparing this variation Va with a threshold Sa such that:
    • when Va is greater than the threshold Sa, the value of N 2 Δ is validated as shock and the values N 2 , N 2 Δ are used in a mathematical function integrating a threshold and a derivation of Va in order to detect an exceeding of the threshold followed by a return to a normal state in a determined time characteristic of a fall giving a value of relevance and fall itself compared to a threshold Sa to characterize the fall,
    • when there is less than the threshold Sa the fall is not suspected, when y is higher than the threshold, the fall is suspected.

Avantageusement, le procédé comporte une détection de séquences perte de gravité, retour de gravité, perte de gravité en sorte de filtrer des mouvements opposés à la direction générale d'une chute.Advantageously, the method comprises a detection of gravity loss sequences, gravity feedback, loss of gravity so as to filter movements opposed to the general direction of a fall.

Le procédé peut par ailleurs comporter un filtrage de mouvements répétitifs.The method may further comprise a filtering of repetitive movements.

L'étape de validation ou de non validation peut comporter une série de mesures de pression et un calcul de variation de pression adapté à détecter un relevage du porteur.The validation or non-validation step may comprise a series of pressure measurements and a pressure variation calculation adapted to detect a lift of the wearer.

Selon un mode de réalisation avantageux, le procédé comporte une temporisation avant alarme dite première temporisation.According to an advantageous embodiment, the method includes a delay before alarm called first timer.

Le procédé peut comporter une attente de détection de recouvrement d'un capteur de dévalidation d'alarme pendant ladite première temporisation et une seconde temporisation de validation de durée de recouvrement, l'alarme étant déclenchée si un appui n'est pas détecté pendant la première temporisation ou si ledit recouvrement est d'une durée inférieure à la seconde temporisation.The method may include a recovery detection wait for an alarm disable sensor during said first time delay and a second recovery time validation time delay, the alarm being triggered if a support is not detected during the first timeout. delay or if said overlap is of a duration less than the second time delay.

Selon un mode de réalisation particulier de l'invention, la détection d'un état porté et d'un état non porté du dispositif par un utilisateur est réalisée par mesure de variation du signal I/Q d'un composant radio du dispositif.According to a particular embodiment of the invention, the detection of a worn state and an unworn state of the device by a user is performed by measuring the variation of the I / Q signal of a radio component of the device.

L'invention peut notamment prévoir une modélisation des pas du porteur par l'analyse des petits chocs récurrents dus aux pas qui donne la manière dont il se déplace, ladite modélisation comportant la modélisation des paramètres vitesse de marche, amplitude des pas, régularité, déséquilibre éventuel de la marche et comportant une ou plusieurs étapes de signalisation de variations importantes desdits paramètres susceptibles de traduire un manque d'équilibre pouvant conduire à un risque de chute.The invention may in particular provide a modeling of the wearer's steps by the analysis of the small recurrent shocks due to the steps which gives the manner in which it moves, said modeling comprising the modeling of the parameters speed of walking, amplitude of the steps, regularity, imbalance possible gait and having one or more signaling stages of significant variations of said parameters may reflect a lack of balance that may lead to a risk of falling.

Le procédé de l'invention peut comporter l'activation d'un vibreur à l'issue de l'étape de validation de détection de chute.The method of the invention may include the activation of a vibrator at the end of the fall detection validation step.

L'invention concerne en outre un dispositif portable de surveillance et de détection de chutes d'une personne adapté à mettre en oeuvre le procédé de l'invention et qui comporte un accéléromètre à trois axes pour détecter les chutes, un capteur de désactivation d'alarme, un capteur de pression, un vibreur, des moyens de calcul et des moyens d'alarme.The invention further relates to a portable device for monitoring and detecting falls of a person adapted to implement the method of the invention and which comprises a three-axis accelerometer for detecting falls, a deactivation sensor of alarm, a pressure sensor, a buzzer, calculation means and alarm means.

Le dispositif peut comporter un boîtier étanche alimenté par pile pouvant être porté en en médaillon, en bracelet, ou à la ceinture.The device may include a battery-powered sealed housing that can be worn as a medallion, bracelet, or belt.

Le dispositif peut aussi être miniaturisé pour être incorporé dans un vêtement ou dans tout objet porté.The device can also be miniaturized to be incorporated in a garment or in any item worn.

Les moyens d'alarme peuvent notamment comporter des composants radio adaptés à transmettre les alarmes et les données de fonctionnement du dispositif soit vers un téléphone mobile soit vers une base de téléassistance soit vers un réseau LPWAN (Low Power Wide Area Network).The alarm means may include radio components adapted to transmit the alarms and the operating data of the device either to a mobile phone or to a remote support base or to a LPWAN (Low Power Wide Area Network) network.

Brève description des dessinsBrief description of the drawings

D'autres caractéristiques et avantages de l'invention seront apparents à la lecture de la description qui suit d'un exemple non limitatif de réalisation de l'invention en référence aux dessins qui représentent:

  • En figure 1 : un schéma simplifié d'un dispositif de l'invention;
  • En figure 2: un synoptique de fonctionnement du procédé.
Other characteristics and advantages of the invention will become apparent on reading the following description of a nonlimiting exemplary embodiment of the invention with reference to the drawings which represent:
  • In figure 1 : a simplified diagram of a device of the invention;
  • In figure 2 : a synoptic of operation of the process.

Description détaillée de modes de réalisation de l'inventionDETAILED DESCRIPTION OF EMBODIMENTS OF THE INVENTION

Le dispositif et le procédé de la présente invention ont pour objectif premier d'aider au maintien à domicile des personnes âgées en prévenant les risques de chute par analyse des pas, en détectant les chutes si elles adviennent malgré tout et en envoyant des alarmes par radio vers une base reliée à un centre de surveillance ou vers un téléphone mobile pourvu d'une application de télésurveillance en cas de chute ou de comportement anormal. Ils peuvent être aussi utilisés pour la protection des travailleurs isolés ou des sportifs de plein air et toute personne où la situation de chute représente un risque.The device and the method of the present invention have the primary objective of helping to keep elderly people at home by preventing the risks of falling by step analysis, by detecting falls if they happen anyway and by sending alarms by radio. to a base connected to a monitoring center or to a mobile telephone equipped with a remote monitoring application in the event of a fall or abnormal behavior. They can also be used for the protection of lone workers or outdoor sportsmen and anyone who is in danger of falling.

Le dispositif de l'invention comporte un boîtier étanche, semblable à un boîtier de montre par exemple, alimenté par pile ou batterie rechargeable pouvant être porté en médaillon, en bracelet, ou à la ceinture mais pouvant être miniaturisé pour être incorporé dans un vêtement ou dans tout objet porté .The device of the invention comprises a sealed housing, similar to a watch case for example, powered by a rechargeable battery or battery which can be worn as a medallion, a bracelet or a belt but which can be miniaturized to be incorporated into a garment or in any item worn.

Le dispositif est équipé de composants radio 5 et d'une antenne 8 permettant de transmettre les alarmes et les données de fonctionnement du dispositif soit en Bluetooth basse énergie (Bluetooth Low Energy) vers un téléphone mobile par exemple, soit en radio dans la bande ISM 868 Mhz pour le dialogue avec une base de téléassistance ou un réseau LPWAN (Low Power Wide Area Network) par exemple.The device is equipped with radio components 5 and an antenna 8 for transmitting alarms and operating data of the device either in low energy Bluetooth (Bluetooth Low Energy) to a mobile phone for example, or in radio in the ISM band 868 Mhz for the dialogue with a base of tele-assistance or a LPWAN network (Low Power Wide Area Network) for example.

Dans le cas d'une base de téléassistance, cette dernière est disposée à poste fixe dans le bâtiment dans lequel se trouve la personne portant le dispositif de sorte que le dispositif puisse dialoguer avec la base et lui transmettre les informations de détection.In the case of a remote assistance base, the latter is arranged in a fixed position in the building in which the person carrying the device is located so that the device can interact with the base and transmit the detection information.

A l'extérieur du bâtiment le dispositif peut dialoguer soit avec un téléphone mobile soit avec un réseau LPWAN pour transmettre les alarmes.Outside the building, the device can communicate with either a mobile phone or an LPWAN network to transmit alarms.

Le dispositif comporte en particulier un accéléromètre 3 axes pour détecter les chutes, un capteur sensitif 6 pour permettre de désactiver une alarme lorsque la personne le souhaite, un capteur de pression 4 qui va servir à confirmer une chute ou au contraire à l'infirmer et un vibreur 9.The device comprises in particular a 3-axis accelerometer for detecting falls, a sensitive sensor 6 to enable an alarm to be deactivated when the person wishes, a pressure sensor 4 which will serve to confirm a fall or, on the contrary, to invalidate it and a vibrator 9.

Le dispositif comporte des moyens de traitement et de calcul sous forme d'un microprocesseur ou microcontrôleur 1 et des moyens de mémorisation données/programme 2 en mémoire vive et mémoire morte ou reprogrammable pour mémoriser les données et le programme interne. Le dispositif réalise un traitement du signal fourni par l'accéléromètre et des autres moyens de détection et d'alarme.The device comprises processing and calculation means in the form of a microprocessor or microcontroller 1 and data storage means / program 2 in RAM and ROM or reprogrammable to store the data and the internal program. The device performs a signal processing provided by the accelerometer and other means of detection and alarm.

L'ensemble des composants du dispositif sont disposés de manière traditionnelle sur un circuit imprimé et reliés par des pistes électriques sous forme d'un ou plusieurs bus 7 et les moyens de traitement et de calcul comportent un ou plusieurs convertisseurs numérique analogique pour traiter les données de l'accéléromètre et du capteur de pression par exemple.All the components of the device are conventionally arranged on a printed circuit and connected by electric tracks in the form of one or more buses 7 and the processing and calculation means comprise one or more digital-to-analog converters for processing the data. accelerometer and pressure sensor for example.

Le dispositif est alimenté par une pile de type pile bouton ou une batterie rechargeable.The device is powered by a button-type battery or a rechargeable battery.

Plusieurs sécurités sont présentes pour détecter que le dispositif est bien porté et pour valider ou invalider les détections de chute.Several safeguards are present to detect that the device is well worn and to validate or invalidate the fall detections.

La vérification du port effectif de l'objet se fait en étudiant la variation du signal I/Q donné par le composant radio 5. En effet la présence d'un corps humain à proximité du dispositif va modifier le signal I/Q en modifiant l'accord de l'antenne 8 du dispositif ce qui permet de détecter si le dispositif est porté ou non.Verification of the actual port of the object is done by studying the variation of the I / Q signal given by the radio component 5. Indeed the presence of a human body in the vicinity of the device will modify the I / Q signal by changing the tuning of the antenna 8 of the device which can detect whether the device is worn or not.

L'alarme chute est signalée au porteur par le vibreur 9 et le porteur peut éventuellement la désactiver en mettant la main sur le capteur sensitif 6 avant que l'alarme soit transmise.The fall alarm is signaled to the carrier by the vibrator 9 and the carrier can possibly disable it by putting your hand on the sensor sensor 6 before the alarm is transmitted.

Le dispositif comporte en outre deux témoins lumineux, diodes led 11 rouge pour signaler une détection de chute et/ou l'envoi d'une alarme, diode led 12 verte liée au fonctionnement et un bouton poussoir 13 permettant de déclencher manuellement une alarme.The device further comprises two LEDs 11, red LEDs to signal a fall detection and / or the sending of an alarm, a green LED 12 linked to the operation and a push button 13 for manually triggering an alarm.

Pour minimiser au maximum les fausses détections de chute, l'alarme est annulée automatiquement si après l'événement interprété comme une chute:

  • Des pas sont détectés,
  • L'altitude du dispositif augmente d'une hauteur correspondant au relevé d'une personne, par exemple plus de 60 cm,
  • Le capteur sensitif du dispositif détecte les événements « recouvert par une main » pendant 3s puis « découvert »,
    ou si le dispositif n'était pas porté par son utilisateur.
To minimize false fall detections, the alarm is canceled automatically if after the event interpreted as a fall:
  • Steps are detected,
  • The altitude of the device increases by a height corresponding to the reading of a person, for example more than 60 cm,
  • The sensory sensor of the device detects events "covered by a hand" for 3s then "discovered",
    or if the device was not worn by its user.

Dans le cas d'un portage en pendentif, un apprentissage de la position pseudo-verticale est mis en place visant à réduire les fausses alarmes tant que ce modèle d'orientation est valide.In the case of a pendant port, a pseudo-vertical position is set up to reduce false alarms as long as this orientation model is valid.

En effet lorsque le dispositif est porté en pendentif, la modélisation de la position peut être réalisée puisque le dispositif maintient une pseudo verticalité. Un calcul de variance sur X, Y, Z sur une période de l'ordre de la seconde caractérise la stabilité de l'orientation dans l'espace du porteur. Un modèle de cette orientation est établi et mis à jour à partir des moyennes de X, Y, Z, en ne retenant que les orientations suffisamment stables. Le suivi de ce modèle d'orientation permet d'isoler les variations d'orientation limitées, même entrecoupées de brèves variations rapides, qui correspondent à une activité maîtrisée. Cette détection permet d'infirmer une chute, si l'orientation du dispositif respecte le modèle établi, ou de la confirmer si le modèle n'est plus validé.Indeed, when the device is worn as a pendant, the modeling of the position can be performed since the device maintains a pseudo verticality. A variance calculation on X, Y, Z over a period of the order of one second characterizes the stability of the orientation in the space of the wearer. A model of this orientation is established and updated from the averages of X, Y, Z, retaining only sufficiently stable orientations. The follow-up of this orientation model makes it possible to isolate limited orientation variations, even interspersed with brief rapid variations, which correspond to a controlled activity. This detection makes it possible to cancel a fall, if the orientation of the device respects the established model, or to confirm it if the model is no longer validated.

En prévention, le dispositif va modéliser les pas du porteur par l'analyse, au moyen de l'accéléromètre 3 des petits chocs récurrents dus aux pas qui donne la manière dont il se déplace et en particulier, la vitesse de marche, l'amplitude des pas, leur régularité, le déséquilibre éventuel de la marche et signaler des variations importantes susceptibles de traduire un manque d'équilibre pouvant conduire à un risque de chute.In prevention, the device will model the wearer's steps by analyzing, using the accelerometer 3 small recurring shocks due to the steps that gives the manner in which it moves and in particular, the walking speed, the amplitude steps, their regularity, the possible imbalance of the march and to announce important variations likely to translate a lack of equilibrium which could lead to a risk of falling.

La détection de chute elle même est réalisée par analyse des mesures acquises par l'accéléromètre 3.The fall detection itself is performed by analyzing the measurements acquired by the accelerometer 3.

L'accéléromètre transmet une mesure combinant l'accélération due à la gravité et l'accélération due aux mouvements du porteur du dispositif. Pour minimiser les calculs, les données d'accélérométrie dans les 3 axes (X,Y,Z) de l'accéléromètre sont principalement travaillées en norme2 c'est-à-dire sous la forme X2+Y2+Z2.The accelerometer transmits a measurement combining the acceleration due to gravity and the acceleration due to the movements of the wearer of the device. To minimize the calculations, the accelerometer data in the 3 axes (X, Y, Z) of the accelerometer are mainly worked in standard 2, ie in the form X 2 + Y 2 + Z 2 .

Il est à noter que les axes X, Y et Z de l'accéléromètre sont fixes par rapport à l'accéléromètre dont la position dans l'espace est quelconque.It should be noted that the X, Y and Z axes of the accelerometer are fixed relative to the accelerometer whose position in space is arbitrary.

Les mesures données par l'accéléromètre pour les 3 axes sont une combinaison de la composante gravité et des composantes mouvement.The measurements given by the accelerometer for the 3 axes are a combination of the gravity component and motion components.

Le principe de détection d'une chute de la présente invention est schématisé à la figure 2. Il consiste à analyser les mesures au travers de trois prismes d'étude: la recherche de la chute libre du dispositif 100 correspondant à une annulation de la gravité qui est symptomatique d'une chute libre, l'analyse des mouvements 200 de la personne, du fait que la chute d'une personne s'accompagne de mouvements, et la détection de chocs 300 par détection des rebonds de la masselotte de l'accéléromètre et la variation de l'angle du vecteur représenté par X,Y et Z dans le temps au moyen d'un calcul de l'angle solide formé entre 2 échantillons temporellement rapprochés, cette détection étant nécessaire pour déterminer une fin de chute.The fall detection principle of the present invention is schematized in the figure 2 . It consists in analyzing the measurements through three study prisms: the search for the free fall of the device 100 corresponding to a cancellation of the gravity which is symptomatic of a free fall, the analysis of the movements 200 of the person, the fact that the fall of a person is accompanied by movements, and the detection of shocks 300 by detection of the rebound of the accelerometer weight and the variation of the angle of the vector represented by X, Y and Z in the time by means of a computation of the solid angle formed between 2 temporally close samples, this detection being necessary to determine an end of fall.

Pour caractériser un évènement de chute, on va filtrer les données et poser des seuils pour les 3 paramètres: perte de la composante gravité, intensité du mouvement, détection de chocs puis combiner les résultats après filtre et seuils au travers d'une fonction mathématique dont le résultat sera lui aussi comparé à un seuil pour déterminer s'il y a chute ou non.To characterize a fall event, we will filter the data and set thresholds for the 3 parameters: loss of the gravity component, intensity of the movement, detection of shocks and then combine the results after filter and thresholds through a mathematical function of which the result will also be compared to a threshold to determine whether there is a fall or not.

Pour la détection, trois séries de calculs sont réalisées.For detection, three sets of calculations are performed.

Une première série 100 de calculs correspond à la recherche de la chute libre et comprend un calcul 101 de la norme au carré N2 des positions X2+Y2+Z2, un stockage 102 de n1 valeurs de N2.A first series of computations corresponds to the search for the free-fall and comprises a calculation 101 of the norm squared N 2 of the positions X 2 + Y 2 + Z 2 , a storage 102 of n1 values of N 2 .

Ces n1 valeurs vont être traités suivant deux branches :

  • Une première branche comprend un calcul 103 de a valeur moyenne des n1 valeurs de N2, le calcul de b valeur moyenne des 4 dernières valeurs de N2, Le calcul (104) de c=a-b et un filtrage 105;
  • Une seconde branche comprend la comparaison (106) des valeurs successives de n2 échantillons de la norme au carré N2 par rapport à un seuil, si n2 échantillons consécutifs sont inférieurs au seuil un indicateur de Chute libre est positionné.
These n1 values will be treated according to two branches:
  • A first branch comprises a calculation 103 of a mean value of the n 1 values of N 2 , the calculation of b average value of the last 4 values of N 2 , the calculation (104) of c = ab and a filtering 105;
  • A second branch comprises comparing (106) the successive values of n2 samples of the norm squared N 2 with respect to a threshold, if n 2 consecutive samples are below the threshold a free fall indicator is set.

Une deuxième série 200 de calculs correspond à l'analyse du mouvement et comporte un stockage 201 de n3 échantillons de X, Y, Z, le calcul 202 de la différence Δ entre deux échantillons, le calcul 203 de la norme au carré du Δ: N2Δ = ΔX2+ ΔY2+ ΔZ2, un filtrage 204 des valeurs N2Δ.A second series 200 of computations corresponds to the motion analysis and comprises a storage 201 of n 3 samples of X, Y, Z, the calculation 202 of the difference Δ between two samples, the calculation 203 of the norm squared of the Δ : N 2 Δ = ΔX 2 + ΔY 2 + ΔZ 2 , a filter 204 of the values N 2 Δ.

A l'étape 205 on prend en compte l'indicateur chute libre de l'étape 106 qui, si il est positionné, va mettre la valeur N2Δ à une valeur maximale prédéterminée. A l'étape 206 est effectuée une comparaison de N2Δ avec un seuil SΔ.In step 205, the free-fall indicator of step 106 is taken into account, which, if it is set, will set the value N 2 Δ to a predetermined maximum value. In step 206 is performed a comparison of N 2 Δ with a threshold S Δ .

Une troisième série 300 de calculs correspond à la détection de chocs et comporte un stockage 301 de n4 valeurs de X, Y, Z, le calcul 302 de la norme au carré X2+Y2+Z2, le calcul 303 de la variation de l'angle Va entre les échantillons et la comparaison 304 de cette variation Va à un seuil Sa.A third series 300 of calculations corresponds to the detection of shocks and comprises a storage 301 of n 4 values of X, Y, Z, the calculation 302 of the standard squared X 2 + Y 2 + Z 2 , the calculation 303 of the variation of the angle V a between the samples and the comparison 304 of this variation V a with a threshold S a .

Lorsque Va est supérieur au seuil Sa, la valeur de N2Δ est validée en 207 et les valeurs N2, N2Δ sont utilisées dans une fonction mathématique 400 donnant une valeur de pertinence de chute elle même comparée à un seuil 410 pour caractériser la chute.When Va is greater than the threshold Sa, the value of N 2 Δ is validated at 207 and the values N 2 , N 2 Δ are used in a mathematical function 400 giving a drop relevance value itself compared to a threshold 410 to characterize the fall.

La fonction mathématique, qui prend en compte la valeur filtrée à l'étape 105 de c et la norme carrée du Δ (N2Δ) validée (207), intègre notamment une multiplication de ces 2 valeurs (y1), la soustraction d'une constante au résultat de la multiplication pour éliminer les évènements mineurs (y2).The mathematical function, which takes into account the value filtered in step 105 of c and the square standard of Δ (N 2 Δ) validated (207), integrates in particular a multiplication of these 2 values (y 1 ), the subtraction of a constant to the multiplication result to eliminate minor events (y 2 ).

Cette fonction permet de détecter un dépassement du seuil suivi d'un retour à un état normal dans un temps déterminé caractéristique d'une chute ou a minima de la phase terminale d'une chute (avant arrêt même bref lorsqu'il est suivi de convulsions)
Le principe d'analyse des données en norme au carré oblige à prendre en compte un certain nombre de phénomènes et les filtrer. En particulier l'étirement d'échelle du bruit (amplifié par le carré), la récupération des valeurs signées lorsque cela s'avère nécessaire
Pour la détection de la disparition de la composante gravité il faut traiter le cas où, pendant la chute, un mouvement du bras, dans le sens de la chute par exemple, va faire que l'accéléromètre va détecter une accélération qui peut être interprétée comme un retour de la composante gravité alors qu'on est bien en face d'une chute. Pour cela on va isoler les séquences, perte de gravité, retour de la gravité, perte de la gravité.This function makes it possible to detect an exceeding of the threshold followed by a return to a normal state in a determined time characteristic of a fall or at least the end phase of a fall (before even stopping briefly when it is followed by convulsions )
The principle of squared standard data analysis makes it necessary to take into account a certain number of phenomena and to filter them. In particular the scaling of noise (amplified by the square), the recovery of signed values when necessary
For the detection of the disappearance of the gravity component it is necessary to treat the case where, during the fall, a movement of the arm, in the direction of the fall for example, will make that the accelerometer will detect an acceleration which can be interpreted as a return of the gravity component while we are in front of a fall. For this we will isolate the sequences, loss of gravity, return of gravity, loss of gravity.

De même, sur l'analyse du mouvement il faut filtrer les mouvements répétitifs qui bien évidemment ne correspondent pas à une chute. Ce filtrage intègre une autocorrélation des signaux X, Y, Z pris séparément sur différentes échelles de temps afin de détecter plusieurs spectres de fréquences caractéristiques par ex dans une gamme de 0.3 à 3 HzIn the same way, on the analysis of the movement it is necessary to filter the repetitive movements which obviously do not correspond to a fall. This filtering integrates an autocorrelation of the signals X, Y, Z taken separately on different time scales to detect several characteristic frequency spectra eg in a range of 0.3 to 3 Hz

Le dispositif acquiert les données de l'accéléromètre en X, Y, Z entre 20Hz et 30Hz et plus particulièrement autour de 25Hz et l'accéléromètre comporte un filtre passe bas entre 1000Hz et 2000Hz, typiquement à 1500Hz.The device acquires the data of the accelerometer in X, Y, Z between 20Hz and 30Hz and more particularly around 25Hz and the accelerometer comprises a low pass filter between 1000Hz and 2000Hz, typically at 1500Hz.

Dans le cas d'un système avec une base de télésurveillance, cette dernière est équipée de moyens de connexion au réseau téléphonique, de moyens radios compatibles avec les moyens radios du dispositif, de moyens de traitement et de calcul adaptés à réaliser une gestion des alarmes et une surveillance du dispositif. Le dispositif est optimisé pour avoir une faible consommation ce qui est possible en limitant l'usage d'un capteur de pression uniquement dans des phases de recherche d'une remontée du dispositif pour invalider une alarme.In the case of a system with a remote monitoring base, the latter is equipped with connection means to the telephone network, radio means compatible with the radio means of the device, processing and calculation means adapted to perform an alarm management and monitoring the device. The device is optimized to have a low power consumption which is possible by limiting the use of a pressure sensor only in the research phase of a recovery of the device to invalidate an alarm.

Dans le cas d'une détection de chute suivie d'un relevage, le dispositif peut éventuellement envoyer une information spécifique qui peut par exemple d'alerte mineure pour le personnel de surveillance.In the case of a fall detection followed by a lifting, the device can possibly send specific information which can for example of minor alert for the surveillance personnel.

La combinaison de la mesure en norme au carré avec des mesures de détection de chocs et d'analyse de mouvement améliorent sensiblement le filtrage des fausses détections de chute par rapport aux dispositifs antérieurs à capteurs accéléromètres. L'invention n'est pas limitée à l'exemple représenté et notamment le dispositif dans une version plus complexe peut comporter d'autres capteurs comme notamment un capteur GPS.The combination of squared standard measurement with shock detection and motion analysis measures substantially improves the filtering of false fall detections compared to prior devices with accelerometer sensors. The invention is not limited to the example shown and in particular the device in a more complex version may include other sensors such as a GPS sensor.

Un mode de veille par détection d'inactivité sur quelques secondes vient compléter le système, en réduisant la consommation énergétique. La phase d'endormissement et de réveil gère toutes les mises à jour et invalidation nécessaires des éléments mémorisés.A standby mode of inactivity detection for a few seconds completes the system, reducing energy consumption. The sleep and wake-up phase manages all the necessary updates and invalidation of the memorized elements.

Claims (13)

  1. Method for monitoring the condition of a person wearing a device for monitoring and detecting falls, carried out by analysing measurements acquired by a triple-axis accelerometer which transmits a measurement combining an acceleration attributable to gravity and an acceleration attributable to the movements of said person, the method comprising:
    - carrying out a free fall search calculation (100) comprising calculating (101) the square norm N2= X2+ Y2+ Z2 of successive measurement data samples of the axes X, Y and Z of the accelerometer, storing (102) n1 values of N2, calculating (103) a mean value a of the n1 values of N2 and a mean value b of the p last values of N2, calculating (104) the difference c=a-b, filtering (105);
    - comparing (106) successive values of the square norm N2 relative to a threshold for positioning a free fall indicator if n2 successive values of N2 are less than the threshold; said method being characterised in that it comprises:
    - carrying out a movement analysis calculation (200) comprising storing (201) n3 samples of X, Y, Z, calculating (202) the difference Δ between samples close to one another in time, calculating (203) the square norm of said difference Δ: N2Δ= ΔX2 + ΔY2 + ΔZ2, filtering (204) the values N2Δ, replacing (205) the calculated values N2Δ with a predetermined value if the free fall indicator has been positioned, comparing (206) N2Δ with a threshold SΔ;
    - carrying out an impact detection calculation (300) comprising storing (301) n4 values of X, Y, Z, calculating (302) the square norm N2= X2+ Y2+ Z2, calculating (303) the angle variation Va between the samples by calculating the solid angle adapted to the square norms;
    - a step of comparing (304) this variation Va with a threshold Sa such that:
    -- if Va is greater than the threshold Sa, the value of N2Δ is confirmed (207) as an impact and the values N2, N2Δ are used in a mathematical function integrating a threshold and a differentiation of Va in order to detect whether the threshold was exceeded followed by a return to a normal condition within a determined period of time that is characteristic of a fall (400) giving a fall relevancy value y itself compared to a threshold Sa (410) in order to characterise the fall,
    -- if y is less than the threshold Sa, a fall is not suspected, if y is greater than the threshold, a fall is suspected.
  2. Method according to claim 1, comprising a detection of loss of gravity-return of gravity-loss of gravity sequences so as to filter out movements opposing the general direction of a fall.
  3. Method according to either claim 1 or claim 2, comprising the filtering of repetitive movements.
  4. Method according to claim 1, claim 2 or claim 3, wherein the step of confirming a fall comprises a series of pressure measurements and a calculation of the pressure variation suitable for detecting the wearer getting back up, opposing the confirmation of the fall.
  5. Method according to any of the previous claims, comprising a pre-alarm time delay referred to as a first time delay.
  6. Method according to claim 5, comprising a waiting period of an alarm-cancelling sensor (6, 13) for detecting a recovery during said first time delay and a second time delay for confirming the recovery duration, the alarm being triggered if no press is detected during the first time delay or if said recovery does not last as long as the second time delay.
  7. Method according to any of the previous claims, wherein the detection of a worn state and of a non-worn state of the device by a user is carried out by measuring the variation in the signal I/Q of a radio component (5) of the device.
  8. Method according to any of the previous claims, comprising a modelling of the steps of the wearer by analysing small recurring impacts attributable to steps taken, which gives the manner in which the wearer is moving, said modelling comprising the modelling of the walking speed parameters, step length, regularity, possible walking imbalance and comprising one or more steps of signalling significant variations in said parameters capable of representing a loss of balance that could lead to a risk of falling.
  9. Method according to any of the previous claims, wherein the step of confirming the detection of a fall activates a vibrating mechanism (9).
  10. Wearable device for monitoring and detecting an individual's falls, capable of implementing the method of any of the previous claims, characterised in that it comprises a triple-axis accelerometer (3) for detecting falls, an alarm deactivation sensor (6), a pressure sensor (4), a vibrating mechanism (9), calculation means (1, 2) and alarm means (5, 8).
  11. Device according to claim 10, comprising a sealed casing, powered by a battery and capable of being worn as a pendant, bracelet or on a belt.
  12. Device according to claim 10, characterised in that it is miniaturised in order to be incorporated into an item of clothing or into any worn object.
  13. Device according to claim 10, claim 11 or claim 12, wherein the alarm means comprise radio components (5) suitable for transmitting the alarms and operating data from the device to a mobile telephone, to a remote assistance base, or to an LPWAN (Low Power Wide Area Network).
EP17203320.1A 2016-11-29 2017-11-23 Improved fall detection device Active EP3327690B1 (en)

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
FR1661635A FR3059454B1 (en) 2016-11-29 2016-11-29 METHOD AND DEVICE FOR MONITORING AND DETECTING FALLS

Publications (2)

Publication Number Publication Date
EP3327690A1 EP3327690A1 (en) 2018-05-30
EP3327690B1 true EP3327690B1 (en) 2019-06-05

Family

ID=57796687

Family Applications (1)

Application Number Title Priority Date Filing Date
EP17203320.1A Active EP3327690B1 (en) 2016-11-29 2017-11-23 Improved fall detection device

Country Status (3)

Country Link
EP (1) EP3327690B1 (en)
ES (1) ES2739480T3 (en)
FR (1) FR3059454B1 (en)

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113793097B (en) * 2021-09-07 2024-04-26 珠海格力电器股份有限公司 Analysis method and system for cargo transportation risk

Family Cites Families (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102393992A (en) * 2011-03-15 2012-03-28 苏州摩多物联科技有限公司 Alarm terminal capable of automatically detecting tumble of human body
US9402568B2 (en) * 2011-08-29 2016-08-02 Verizon Telematics Inc. Method and system for detecting a fall based on comparing data to criteria derived from multiple fall data sets
CN102610056B (en) * 2012-03-16 2013-09-04 清华大学 Mobile phone wearing mode-oriented falling event detection system and method

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
None *

Also Published As

Publication number Publication date
FR3059454A1 (en) 2018-06-01
EP3327690A1 (en) 2018-05-30
FR3059454B1 (en) 2019-05-10
ES2739480T3 (en) 2020-01-31

Similar Documents

Publication Publication Date Title
CN102048521B (en) System and method for monitoring and preventing tumble
EP2229665B1 (en) Method and apparatus for detecting a critical situation of a subject
WO2017207925A1 (en) Connected device for behavioural monitoring of an individual and for detecting and/or preventing an anomaly
FR3047574B1 (en) INTELLIGENT AND COMMUNICABLE CLOTHING APPARATUS, METHOD AND INSTALLATION FOR BIDIRECTIONAL COMMUNICATION WITH SUCH A DRESSING ARTICLE
KR100873495B1 (en) Fall detection device and method and fall rescue service system and method using same
FR2981561A1 (en) METHOD FOR DETECTING MOTION SENSOR ACTIVITY, CORRESPONDING DEVICE AND COMPUTER PROGRAM
CA2939633A1 (en) Activity recognition using accelerometer data
Kerdegari et al. A pervasive neural network based fall detection system on smart phone
WO2019063925A1 (en) Device forming an item of personal protective equipment for persons moving in an electrical risk area
US20230252881A1 (en) Systems, devices and methods for fall detection
EP3327690B1 (en) Improved fall detection device
WO2016075013A1 (en) Method for detecting the fall of a human subject and corresponding actimetric device
EP3158343B1 (en) Energy-efficient home-automation device and method for tracking the displacement of a monitored object
EP2207154B1 (en) Method and device for detecting a fall
Arcaya-Jordan et al. Smart Connected Glasses for Drowsiness Detection: a System-Level Modeling Approach
Büsching et al. Fall detection on the road
JP5371103B2 (en) Reporting system, reporting device, reporting control method, and program
FR2776101A1 (en) Surveillance system for detecting presence of objects within monitored zone around control
KR102451630B1 (en) Impairment detection with environmental considerations
WO2022084154A1 (en) Multifunctional sensor for monitoring premises and methods using such a sensor
WO2014098765A1 (en) Medical monitoring systems and methods for using a medical monitoring system
CN112386249A (en) Fall detection method and device, equipment and storage medium
CN205339294U (en) Printed circuit board and have mountain -climbing goggles of health monitoring effect
US20160371951A1 (en) Low power system and device for detecting uv radiation
EP3686858B1 (en) System for monitoring daily activities for one or more persons in a living space, and analysis method

Legal Events

Date Code Title Description
PUAI Public reference made under article 153(3) epc to a published international application that has entered the european phase

Free format text: ORIGINAL CODE: 0009012

STAA Information on the status of an ep patent application or granted ep patent

Free format text: STATUS: THE APPLICATION HAS BEEN PUBLISHED

AK Designated contracting states

Kind code of ref document: A1

Designated state(s): AL AT BE BG CH CY CZ DE DK EE ES FI FR GB GR HR HU IE IS IT LI LT LU LV MC MK MT NL NO PL PT RO RS SE SI SK SM TR

AX Request for extension of the european patent

Extension state: BA ME

STAA Information on the status of an ep patent application or granted ep patent

Free format text: STATUS: REQUEST FOR EXAMINATION WAS MADE

17P Request for examination filed

Effective date: 20181115

RBV Designated contracting states (corrected)

Designated state(s): AL AT BE BG CH CY CZ DE DK EE ES FI FR GB GR HR HU IE IS IT LI LT LU LV MC MK MT NL NO PL PT RO RS SE SI SK SM TR

GRAP Despatch of communication of intention to grant a patent

Free format text: ORIGINAL CODE: EPIDOSNIGR1

STAA Information on the status of an ep patent application or granted ep patent

Free format text: STATUS: GRANT OF PATENT IS INTENDED

INTG Intention to grant announced

Effective date: 20190104

GRAS Grant fee paid

Free format text: ORIGINAL CODE: EPIDOSNIGR3

GRAA (expected) grant

Free format text: ORIGINAL CODE: 0009210

STAA Information on the status of an ep patent application or granted ep patent

Free format text: STATUS: THE PATENT HAS BEEN GRANTED

AK Designated contracting states

Kind code of ref document: B1

Designated state(s): AL AT BE BG CH CY CZ DE DK EE ES FI FR GB GR HR HU IE IS IT LI LT LU LV MC MK MT NL NO PL PT RO RS SE SI SK SM TR

REG Reference to a national code

Ref country code: GB

Ref legal event code: FG4D

Free format text: NOT ENGLISH

REG Reference to a national code

Ref country code: CH

Ref legal event code: EP

REG Reference to a national code

Ref country code: AT

Ref legal event code: REF

Ref document number: 1140780

Country of ref document: AT

Kind code of ref document: T

Effective date: 20190615

REG Reference to a national code

Ref country code: IE

Ref legal event code: FG4D

Free format text: LANGUAGE OF EP DOCUMENT: FRENCH

REG Reference to a national code

Ref country code: DE

Ref legal event code: R096

Ref document number: 602017004328

Country of ref document: DE

REG Reference to a national code

Ref country code: NL

Ref legal event code: FP

REG Reference to a national code

Ref country code: SE

Ref legal event code: TRGR

REG Reference to a national code

Ref country code: LT

Ref legal event code: MG4D

PG25 Lapsed in a contracting state [announced via postgrant information from national office to epo]

Ref country code: NO

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20190905

Ref country code: FI

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20190605

Ref country code: AL

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20190605

Ref country code: LT

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20190605

Ref country code: HR

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20190605

PG25 Lapsed in a contracting state [announced via postgrant information from national office to epo]

Ref country code: GR

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20190906

Ref country code: LV

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20190605

Ref country code: BG

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20190905

Ref country code: RS

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20190605

REG Reference to a national code

Ref country code: AT

Ref legal event code: MK05

Ref document number: 1140780

Country of ref document: AT

Kind code of ref document: T

Effective date: 20190605

PG25 Lapsed in a contracting state [announced via postgrant information from national office to epo]

Ref country code: CZ

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20190605

Ref country code: RO

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20190605

Ref country code: AT

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20190605

Ref country code: EE

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20190605

Ref country code: PT

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20191007

Ref country code: SK

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20190605

REG Reference to a national code

Ref country code: ES

Ref legal event code: FG2A

Ref document number: 2739480

Country of ref document: ES

Kind code of ref document: T3

Effective date: 20200131

PG25 Lapsed in a contracting state [announced via postgrant information from national office to epo]

Ref country code: SM

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20190605

Ref country code: IS

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20191005

REG Reference to a national code

Ref country code: DE

Ref legal event code: R097

Ref document number: 602017004328

Country of ref document: DE

PG25 Lapsed in a contracting state [announced via postgrant information from national office to epo]

Ref country code: TR

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20190605

PLBE No opposition filed within time limit

Free format text: ORIGINAL CODE: 0009261

STAA Information on the status of an ep patent application or granted ep patent

Free format text: STATUS: NO OPPOSITION FILED WITHIN TIME LIMIT

PG25 Lapsed in a contracting state [announced via postgrant information from national office to epo]

Ref country code: DK

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20190605

Ref country code: PL

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20190605

26N No opposition filed

Effective date: 20200306

PG25 Lapsed in a contracting state [announced via postgrant information from national office to epo]

Ref country code: SI

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20190605

PG25 Lapsed in a contracting state [announced via postgrant information from national office to epo]

Ref country code: MC

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20190605

Ref country code: LU

Free format text: LAPSE BECAUSE OF NON-PAYMENT OF DUE FEES

Effective date: 20191123

PG25 Lapsed in a contracting state [announced via postgrant information from national office to epo]

Ref country code: IE

Free format text: LAPSE BECAUSE OF NON-PAYMENT OF DUE FEES

Effective date: 20191123

PG25 Lapsed in a contracting state [announced via postgrant information from national office to epo]

Ref country code: CY

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20190605

PG25 Lapsed in a contracting state [announced via postgrant information from national office to epo]

Ref country code: MT

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20190605

Ref country code: HU

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT; INVALID AB INITIO

Effective date: 20171123

PG25 Lapsed in a contracting state [announced via postgrant information from national office to epo]

Ref country code: MK

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20190605

P01 Opt-out of the competence of the unified patent court (upc) registered

Effective date: 20230525

PGFP Annual fee paid to national office [announced via postgrant information from national office to epo]

Ref country code: NL

Payment date: 20231121

Year of fee payment: 7

PGFP Annual fee paid to national office [announced via postgrant information from national office to epo]

Ref country code: GB

Payment date: 20231128

Year of fee payment: 7

PGFP Annual fee paid to national office [announced via postgrant information from national office to epo]

Ref country code: ES

Payment date: 20231219

Year of fee payment: 7

PGFP Annual fee paid to national office [announced via postgrant information from national office to epo]

Ref country code: SE

Payment date: 20231123

Year of fee payment: 7

Ref country code: IT

Payment date: 20231121

Year of fee payment: 7

Ref country code: FR

Payment date: 20231123

Year of fee payment: 7

Ref country code: DE

Payment date: 20231121

Year of fee payment: 7

Ref country code: CH

Payment date: 20231201

Year of fee payment: 7

PGFP Annual fee paid to national office [announced via postgrant information from national office to epo]

Ref country code: BE

Payment date: 20231123

Year of fee payment: 7