Posture assistance system in a vehicle seat
The system addresses the inflexibility of existing posture correction systems by using sensors to calculate and adaptively correct occupant posture, enhancing safety and compliance through personalized alerts.
Patent Information
- Application Number
- FR2022011316
- Authority / Receiving Office
- FR · FR
- Patent Type
- Patents
- Current Assignee / Owner
- Filing Date
- 2022-10-28
- Publication Date
- 2024-11-08
- Estimated Expiration
- 2042-10-28
AI Technical Summary
Existing vehicle seat posture correction systems lack flexibility and adaptability, often leading to user resistance and potentially dangerous situations due to rigid, binary posture evaluations.
A system that uses a set of sensors to detect occupant posture, calculates multiple comfort indices over time, combines them into an overall index, and generates alerts based on reference values, allowing for personalized and context-aware posture corrections.
Enhances user compliance and safety by providing adaptive, personalized posture corrections, reducing the risk of discomfort, fatigue, and accidents by offering detailed and timely alerts.
Smart Images

Figure 00000018_0000 
Figure 00000018_0001 
Figure 00000019_0000
Abstract
Description
Title of the invention: Posture assistance system in a vehicle seat Technical field
[0001] This disclosure falls within the domain of vehicle embedded systems. Prior art
[0002] It is known to equip vehicle seats with sensors designed to determine information about the posture of their occupants, in particular to detect postures inappropriate to a given situation and correct them quickly. Application FR2101745 filed on February 23, 2021, on behalf of the applicant, describes, for example, a solution for monitoring the trunk rotation of a vehicle occupant.
[0003] Correcting the posture of a vehicle seat occupant generally aims to ensure the proper functioning of safety systems in the event of an incident, such as a seat belt and airbags in a motor vehicle, for example. It also helps the occupant avoid a posture that could, in the long term, lead to the development of disorders or pain. When the seat occupant is also one of the vehicle's drivers, limiting the occurrence of disorders, or even simply fatigue, improves the safety of all vehicle occupants and those nearby.
[0004] However, some known systems are designed in a rigid and binary manner, in that one set of postures is always considered "compliant" while the other postures are always considered "non-compliant." In other words, existing systems may lack flexibility and adaptability to the context of use. When such systems are too restrictive for users, they may try to circumvent the system. This can generate particularly dangerous situations, which is counterproductive. Overly rigid systems can, in certain situations, produce effects contrary to their purpose or give a false sense of security compared to reality. Summary
[0005] The present disclosure improves the situation.
[0006] A method is proposed to assist in correcting the posture of a vehicle seat occupant. The method comprises: a. collect input signals from a set of sensors, said sensors being jointly arranged to detect the posture of the seat occupant; b. For each time step t, calculate, based on the collected signals, at least two comfort indices by implementing at least one algorithm. for at least two of the comfort indices, the time step t is itself a calculation parameter, so that the value of the index evolves according to the posture of the occupant and also according to time; c. for each time step t, calculate an overall comfort index as a combination of all the comfort indices; d. compare each of the comfort indices and the overall comfort index to one or more reference values; e. generate at least one output signal for an alerting device arranged to issue an alert, said generated signal being dependent on the results of the comparison.
[0007] According to another aspect, an embedded vehicle system is proposed comprising: - a seat suitable for accommodating a vehicle occupant; - a set of sensors jointly arranged to detect the posture of the seat occupant, at least some of which are integrated into the seat; - an alert system; and - a controller capable of receiving input signals from the sensor set and generating at least one output signal for the warning device. The controller is also configured to: a. collect input signals from a set of sensors, said sensors being jointly arranged to detect the posture of the seat occupant; b. For each time step t, calculate, based on the collected signals, at least two comfort indices Ci>t by implementing at least one algorithm, for at least two of the comfort indices Ci>t, the time step t being itself a calculation parameter, so that the value of the index Ci>t evolves according to the posture of the occupant and also according to time; c. for each time step t, calculate an overall comfort index CGjt as a combination of all the comfort indices Ci>t; d. compare each of the comfort indices Ci>t and the overall comfort index CGjt to one or more reference values; e. generate at least one output signal for an alerting device arranged to issue an alert, said generated signal being dependent on the results of the comparison.
[0008] According to another aspect, a computer program is proposed comprising instructions for implementing all or part of a process as defined herein when this program is executed by a processor. According to another aspect, a non-transient, computer-readable recording medium is proposed on which such a program is recorded.
[0009] The features described in the following paragraphs may, optionally- In fact, they can be implemented independently of each other or in combination with each other:
[0010] The calculation of at least one of the comfort indices Ci>t is further based on at least one parameter specific to the occupant and previously collected from among anthropometric, physiological, and pathological parameters, an age (or age range), and the occupant's preferences. This allows for further refinement of the relevance of the alerts based on the occupant themselves.
[0011] Each comfort index Ci>t is associated with a part of the occupant's body. This allows for the generation of more detailed and specific alerts, and therefore the communication of more useful information to users.
[0012] The sensor set includes at least one of the following sensors: - a pressure sensor; - a temperature sensor; - a humidity sensor; - an interdigitated sensor, such as a capacitive sensor; - a Hall effect sensor; - a video sensor arranged to capture an image of at least part of the seat occupant and connected to the controller via an image analysis module; - a position and / or orientation sensor of a first part of the seat relative to a second part of the seat; - a position and / or orientation sensor of a part of the seat relative to a part of the vehicle distinct from the seat; - a position and / or orientation sensor for a steering wheel, a seat belt attachment or a vehicle mirror; - a sensor integrated into a seat back; - a sensor integrated into a seat cushion. Brief description of the drawings
[0013] Other features, details and advantages will become apparent upon reading the detailed description below and analyzing the accompanying drawings, in which: Fig. 1
[0014] [Fig.1] schematically represents a system intended to be embedded in a vehicle. Fig. 2
[0015] [Fig.2] shows a diagram of the implementation of a method according to a mode of implementation lisation. Fig. 3
[0016] [Fig.3] shows an example of rendering an alert generated for an occupant of a vehicle seat with a visual representation of an alert indicating an excessively curved, and therefore incorrect, posture of the spine. Fig. 4
[0017] [Fig.4] shows an example of rendering an alert generated for an occupant of a vehicle seat with a visual representation of an alert indicating correct spinal posture. Fig. 5
[0018] [Fig.5] shows an example of rendering an alert generated for an occupant of a vehicle seat with a visual representation of an alert representative of various indices relating to the occupant's posture. Fig. 6
[0019] [Fig. 6] shows an example of rendering an alert generated for a vehicle seat occupant with a visual representation of an overall comfort index in the upper part and several specific indices in the lower part. Description of embodiments
[0020] It is specified that, in the present context and in the context of vehicle seats, the term "comfort" is used in a sense aimed at improving human health and safety, and not solely in its subjective sense of "pleasure" or "enjoyable." In other words, the goal is to improve safety in the field of passenger transport. Improving the occupant's experience of a seat may be an additional, but secondary, benefit. It is also specified that, in the present context, the term "controller" is used in its general sense of a device capable of processing data, such as a system of software components able to interact to implement a computer process.Thus, the controller can, for example, take the form of an integrated circuit (or "ASIC" for "Application-Specific Integrated Circuit"), an electronic control unit (or "ECU" for "Electronic Control Unit"), a processor (or "Central Processing Unit" or "CPU" for "Central Processing Unit"), or a combination of several of the aforementioned elements. In particular, the controller can be an independent component or be a function integrated into an on-board computer among other functions implemented by the on-board computer. The controller is therefore defined here by its function and what it is configured for, rather than by its physical structure.
[0021] Reference is now made to [Fig. 1]. [Fig. 1] represents a system 1 embedded in a vehicle, or intended to be embedded in it. In the example described here, system 1 comprises: - one seat 11; - a set of 12 sensors; - an alert device 13; and - a controller 14. The outputs of the sensors 12 are connected to an input of the controller 14 while an output of the controller 14 is connected to the input of the warning device 13.
[0022] The seat 11 is suitable for accommodating a vehicle occupant, for example, a driver, pilot, or passenger. In the example described here, the seat 11 comprises a seat 111, a backrest 112, and a headrest 113. The seat 11 is further equipped with a set of movable parts and adjustment controls, not shown in [Fig. 1], arranged to allow the position and orientation of the seat 11 to be changed relative to its surroundings (the rest of the vehicle), and also to allow the position and orientation of the seat 11's components to be changed relative to each other. Generally, such adjustments allow the configuration of the seat 11 to be modified at will to help the occupant adopt a desired posture.
[0023] The sensors in the sensor set 12 are jointly arranged to detect the posture of the seat occupant. At least some of the sensors are integrated into the seat 11. The nature, number, and position of the sensors may vary from one embodiment to another, particularly depending on the intended use and the trade-offs between the desired accuracy and the manufacturing and implementation costs of such a system. Examples of sensors that may be included in the sensor set 12 are given below.
[0024] A pressure sensor is generally inexpensive. It can be integrated into one or more parts of the seat 11 such as the seat 111, the backrest 112, the headrest 113 and / or an armrest.
[0025] A temperature sensor indirectly allows the occupant's posture to be deduced by detecting the body heat emitted by the occupant. Due to thermal inertia, temperature sensors also make it possible to distinguish the prolonged presence of a part of the occupant's body from a brief positioning or movement. This is valid regardless of any indexing of the sensor measurements with a temporal reference frame. In other words, when the output signal of a temperature sensor exceeds a predetermined temperature value, it is possible to deduce the presence of a part of the occupant's body nearby for a period exceeding a few seconds or minutes, even in the absence of a history of past measurements.
[0026] A humidity sensor makes it possible, in particular, to infer discomfort by detecting abnormally high humidity, typically resulting from the occupant's perspiration. As with a temperature sensor, a humidity sensor facilitates the distinction between the prolonged presence of a part of the occupant's body and a brief positioning or movement.
[0027] Interdigital capacitive sensors, due to their thinness and flexibility, are particularly well-suited for integration into the seat 11, typically under a surface layer, for example, fabric. Interdigital capacitive sensors can be positioned very close to the occupant without causing any inconvenience to said occupant.
[0028] Hall effect sensors are particularly precise and robust, and therefore reliable. Being inherently sensitive to magnetic fields, they can be used, in particular, to measure the relative position (or orientation) between two metal parts of the seat 11 or between a metal part of the seat 11 and a metal part of the vehicle separate from the seat. Measuring such positions makes it possible to deduce information about the occupant's posture.
[0029] A video sensor, different in nature from those in the preceding examples, can be arranged to capture an image of at least part of the occupant of seat 11, for example, the upper part, above the waist. Such a video sensor then includes a hardware component, for example, a camera, and a software component, for example, an image analysis module 15. The camera is then connected to the image analysis module 15 for processing the video signal. The video signal is converted into digital data representing the occupant's posture. The image analysis module 15 can be integrated into the video sensor itself or be separate, for example, forming part of a computer, such as the controller 14 and / or an on-board computer of the vehicle.
[0030] Depending on their integration into the environment, the sensors in the sensor set 12 can perform various functions. Whether of the nature of one of the preceding examples or another, each sensor can be, for example: - a position and / or orientation sensor of a first part of seat 11 relative to a second part of seat 11; - a position and / or orientation sensor of a part of seat 11 relative to a part of the vehicle distinct from seat 11.
[0031] Some of the sensors may also be associated with parts of the vehicle other than the seat 11 to deduce information about the occupant's posture. For example, in the context of a car, the sensor set 12 may include a position and / or orientation sensor for a steering wheel, a seat belt anchor, or a mirror (or "rearview mirror") of the vehicle. Furthermore, some sensors may be integrated into a human-machine interface and take, for example, the form of a physical or virtual button, allowing a user to transmit, through a voluntary action, information relating to their posture in the seat.
[0032] The alert device 13 is capable of emitting an alert perceptible to at least one occupant of the vehicle. Although it is referred to as an alert in the singular, the The alerting device 13 can also be capable of issuing multiple alerts. The alerting device 13 is configured to issue the alert upon receiving a command signal from the controller 14. The alert can be binary, meaning it is issued upon receipt of the command signal and not issued if the command signal is not received. Alternatively, the alert can take on a variety of shapes or a combination of shapes that depend on the received command signal and, therefore, as will be described later, on the results of the computer process implemented by the controller 14. A composite alert thus allows a wide variety of information to be transmitted to the occupant, as will be exemplified below.
[0033] The warning device 13 can generate, for example, a visual alert. The warning device 13 can include, for example, an indicator light or a combination of indicator lights whose illumination corresponds to one or more separate alerts. Alternatively, a visual alert can be generated via a display screen visible to a vehicle occupant. Examples of such displays are shown in [Fig. 3], [Fig. 4], [Fig. 5], and [Fig. 6]. Of course, the content of the display can be supplemented with other information for the vehicle occupants. In particular, the alert can be one piece of information among others displayed on the screen: the alert display is, for example, one component among others that an on-board computer makes available to the occupants.
[0034] In order to limit visual distractions, which is particularly important when the occupant is also a driver or pilot, visual alerts can be supplemented or replaced by other types of alerts, for example audible or haptic. For this purpose, the alert device 13 can, for example: - understand a loudspeaker and be controlled to emit a sound or voice message as an alert; and / or - include one or more vibrators that activate to form an alert.
[0035] It will be understood that the technical function of the warning device 13 is to alert, and that the examples of warning devices 13 above are illustrative and not limiting. The ways in which information or alerts are presented to the vehicle occupants are virtually infinite and are not, as such, the subject of the improvement described here.
[0036] The controller 14 is capable of receiving input signals from the sensor set 12 and of generating at least one output signal for the warning device 13.
[0037] In some embodiments, at least part of the hardware resources of the controller 14 (computing and / or memory) may be separate from and located remotely from the vehicle. In other words, the method described below can be implemented, in part, remotely. The sensor set 12 and the warning device 13 can be connected to the controller 14 by via communication networks. Particularly in use cases where latency is not a critical parameter, implementing a remote controller allows for resource optimization: a single controller can be used in combination with multiple sets of sensors and warning devices installed in multiple vehicles. Such network operation also facilitates the analysis of the data obtained and the improvement of the implemented models.
[0038] Reference is now made to [Fig. 2]. [Fig. 2] is a functional diagram of an embodiment of a method, implemented by a controller, for assisting in correcting the posture of a vehicle seat occupant. In the following, the numerical references used with reference to [Fig. 1] are reused to designate elements that may be similar. In particular, the method can be implemented by the controller 14 described above.
[0039] In a first operation 1001, signals from a set of sensors 12 are collected. Thus, the controller 14 receives the signals from the sensors as input and can use the data carried by these signals to deduce information about the occupant's posture. In the example described here, the signals obtained are raw and processed by the controller 14. Alternatively, intermediate signal processing can be performed upstream of reception by the controller 14. In an embodiment comprising a video sensor as described above, an image analysis module 15 can be activated and process the video signal output from the video sensor upstream of the controller 14.
[0040] When at least one of the signals from the sensor set is a video signal, the method may optionally include, before calculating the comfort indices Ci>t described below: al. perform an image analysis from the video signal in order to convert the video signal into digital data representative of the posture of the seat occupant.
[0041] The use of a video sensor makes it possible to detect a large number of posture-related parameters. The use of such a sensor can therefore replace or supplement many other sensors while providing measurements and data that are sometimes more reliable and precise. Furthermore, such a video sensor is already available in some vehicles for other functions.
[0042] In a second operation 1002, at least two comfort indices Cijt are calculated. Each comfort index Cijt is calculated based on collected signals. Each comfort index Cijt is calculated by implementing at least one algorithm. In the example described below, each comfort index Cijt is calculated by implementing an algorithm different from those implemented for the others. Comfort indices. For each comfort index Ci>t, a value is calculated for each time step t. Furthermore, time t is itself a parameter for calculating the comfort index Ci>t. In other words, the value of the comfort index Ci>t depends both on the occupant's posture as detected by the sensors 12 and on time. Thus, the value of the comfort index Ci>t changes from one time step t to the next t+1 (it is not constant), even if the occupant's posture does not change from time step t to the next t+1, for example, because the occupant is stationary in the seat.
[0043] We now give some examples of indices C that are relevant to calculate and examples of corresponding calculation algorithms. Note that the first two examples that follow are dependent on the passage of time.
[0044] A first example of an index, denoted i = 1, represents the comfort associated with the time spent by the occupant in the seat: Ci,t. Such an index can be constructed to decrease as the occupant spends time in the seat and to be reset when the occupant sits down in the seat. The decrease in the index can be linear with respect to time. The index Ci,t can, for example, be calculated according to the following formula [Math. 1].
[0045] [Math.l] 1 / ~ IJnit ~ F X.( t - tstop)
[0046] Cijnit being a predetermined initial or ideal value of the index, for example each Ci_init value of the indices Ci described here is fixed at 80; F being a constant factor, for example equal to 29 / 5400; t being the current date; tstop being the date of the occupant's installation in the seat, determined based on signals from the sensors.
[0047] A second example of an index, indexed i = 2, represents the comfort associated with the time spent by the occupant in a fixed posture: C2,t. Such an index can be constructed to decrease as the occupant remains fixed (immobile) and to be reset when the occupant changes posture in the seat. The decrease in the index can be linear with respect to time. The index C2jt can, for example, be calculated according to the following formula [Math. 2].
[0048] [Math.2] — Cy init “ FX £ X - tffap)
[0049] C2_init being a predetermined initial value of the index; E being a constant factor, for example equal to 1.6 F; z being the proportion of time spent in a fixed posture relative to the time spent in the seat since the occupant was seated, for example calculated according to equation [Math. 3] below, a fixed posture is distinguished from others based on the signals from the sensors.
[0050] [Math.3] cumulative time spent in a fixed position [Math.3] (MSfUp) - cumulative time spent in motion or in a non-condom posture
[0051] A third example of an index, denoted i = 3, represents the comfort associated with the occupant's pelvic position in the seat, or the discomfort associated with the occupant's pelvic drift in the seat: C3. In the example described here, the index C3 is a discrete value, which varies according to the occupant's pelvic position in the seat. Alternatively, the index C3 can be a Boolean, that is, have a fixed value when the pelvic position is correct and another fixed value otherwise. In the example described here, the index C3 can, for example, be calculated according to the following formula [Math. 4].
[0052] [Math.4]
[0053] C3_init being a predetermined value of the index corresponding to a position of the occupant's pelvis in the seat, considered to be ideal; A being a constant factor, for example equal to 10 / (7200*10); W being a value of the drift of the occupant's pelvis relative to a position considered ideal and determined from the signals from the sensors.
[0054] A fourth example of an index, indexed i = 4, represents the comfort associated with the curvature of the occupant's back in the seat, or the discomfort associated with the curvature of the occupant's back in the seat: C4. In the example described here, the index C4 is a discrete value, which varies according to the curvature of the occupant's back in the seat. Alternatively, the index C4 can be a Boolean, that is, have a fixed value when the curvature of the back is compliant and another fixed value otherwise. In the example described here, the index C4 can, for example, be calculated according to the following formula [Math. 5].
[0055] [Math.5] C4 = C 4Jnit -BxX
[0056] C4_init being a predetermined value of the index corresponding to a curvature of the occupant's back in the seat, considered to be ideal; B being a constant factor, for example equal to 10 / (7200*10); X being a value of the curvature of the occupant's back relative to a curvature considered ideal and determined from the signals from the sensors.
[0057] In the preceding index examples, it is possible to customize the predetermined initial or ideal values Ci_init. In other words, the Ci_init values can be predetermined but specific to the occupant. Thus, data specific to each potential occupant can be collected beforehand and then made available to the controller, including anthropometric, physiological, and pathological parameters, age, and occupant preferences. The Ci_init value is therefore constant for an occupant but can vary from one occupant to another. The same applies to other values used in the calculation of the indices, such as factors A, B, E, and F mentioned above. In this context, anthropometric parameters can include morphological parameters such as weight and / or body dimensions.Physiological parameters include, for example, body temperature, humidity, heart rate and / or an indicator specifying whether the occupant is pregnant or not.
[0058] Of course, the above examples of indices are illustrative and not limiting: other algorithms can be implemented and other comfort indices can be calculated from information derived from sensor signals. In particular, other local indices (indices related to parts of the occupant's body other than the torso) can be calculated, for example, related to the legs, feet, arms and / or neck / head.
[0059] In a third operation 1003, for each time step t, an overall comfort index CGjt is calculated as a combination of comfort indices Ci>t from among those calculated during operation 1002. In the example described here, the overall comfort index CGjt is calculated as a sum, weighted or unweighted, of the indices defined above. The overall comfort index CGjt can, for example, be calculated according to the following formula [Math. 6].
[0060] [Math.6] y” £2
[0061] n being the number of indices involved in the calculation of the overall comfort index, with n = 4 in this example.
[0062] The above calculation formula corresponds to a method of calculating an unweighted sum (or, equivalently, with equal weightings from one index to another). Alternatively, weightings specific to each index can be applied. In particular, the weighting coefficients can be set by default or depend on data specific to the occupant. Such data specific to each potential occupant can be collected beforehand and made available to the controller, including anthropometric, physiological, and pathological parameters, as well as the occupant's preferences. Thus, if an occupant indicates, by Through a human-machine interface, for those particularly sensitive to or prone to back pain, weighting coefficients can be selected to make indices related to back comfort predominant in the calculation of the overall comfort index. The same applies, for example, to occupants whose morphological data corresponds to a tall person relative to the general population.
[0063] In the preceding examples, the indices are calculated using explicit and determinate equations. Alternatively, the calculation of at least one of the comfort indices Cit is also a function of at least one predictive model previously built using machine learning. The construction of such models and their evolution can be facilitated by implementing the processes described here in a centralized manner for a plurality of vehicles. Indeed, the volume of data received by a "central" controller (processing data from a large number of sensor sets) can be used to feed the machine learning database that builds (or corrects) the model.
[0064] In a fourth operation 1004, each of the comfort indices Cijt and the overall comfort index CG,t is compared to one or more reference values Ci>min, Ci>max, Cg,min, CG>max. The reference values are predetermined. The comparison operation aims, for each index and each time step t, to determine whether the value of each index falls within a nominal range considered "normal" or not. Here, a minimum and a maximum value are provided for each index. Alternatively, a single reference value may suffice. It should be noted, however, that depending on the calculation method for each index, specifying a value beyond which "comfort" can be considered "abnormally high" may, in practice, correspond to an undesirable situation, for example, one that promotes reduced attention and responsiveness in a driver, or even leads to drowsiness.
[0065] In a fifth operation 1005, a signal intended for an alerting device 13 is generated at the output (of the controller 14). As explained previously, the signal is configured so that the alerting device 13 emits an alert upon receipt. Such an alert may be silent (intended for other vehicle components) and / or perceptible to at least one vehicle occupant. The generated signal depends on the results of the comparison. For example, a signal may be emitted only when the results of the comparison operation 1004 lead to a situation in which at least one of the indices Cijt, CG,t is outside a nominal range, while no signal is generated as long as each index is within its nominal range. In alternative embodiments, such as that shown in [Fig. 5] or [Fig. 6], for which it is desired to be able to provide more detailed information, the signal The alert is composite: it carries varied information, and is generated even when the indices are within their respective nominal ranges, for example periodically or almost continuously.
[0066] Reference is now made to [Fig. 3] and [Fig. 6], which show examples of a human-machine interface (e.g., a screen) that can act as an alerting device. Although the way the information is presented is purely illustrative and not limiting, the reader will infer variations in implementation. In particular, the use of colors is one possible means, but one that is difficult to illustrate with monochrome figures. This visual example shows, among other things, that it is possible to quickly provide a user with general and easily understandable information, including while driving, in the form of an "overall score" or "tips." It is also possible to provide detailed information on the occupant's posture to help the user better adapt their posture, for example, by distinguishing the relevant body parts.
[0067] This disclosure therefore makes it possible to both facilitate and accelerate posture corrections, and to prevent disorders that can arise from non-compliant posture, including postures for which discomfort or pain is not immediately perceptible to humans. In so doing, when the occupant is also the driver of the vehicle, the risks of driving errors and resulting accidents are also reduced. Industrial application
[0068] The preceding examples are easily understood in that they evoke a context of common use for a large number of readers, namely the position of the driver in their car seat to whom information and alerts are transmitted while driving (without distracting the driver from driving). It will be understood, however, that the present disclosure is not limited to such a context. For example, the position analyzed could be that of a first person, such as a child passenger in the vehicle, while the alerts and information are transmitted to a second person, for example, the driver or any other occupant presumed to be an adult responsible for the child. It should also be noted that the alerts and information could also be transmitted outside the vehicle by means of communication known per se, for example, to emergency services in the event of an accident.Vehicles can also be different from a car, for example trains or aircraft. The context of use, and in particular the length of time the occupant remains in the seat, can therefore be significantly different from the usual duration for a car used by a non-professional user.
[0069] This disclosure is not limited to the examples of processes, systems, computer programs and recording media for such programs described above, only by way of example, but encompasses all the variants that the reader may consider in the context of the protection sought. List of reference signs
[0070] - 1: System - 11: seat; - 12: sensor set; - 13: alert device; - 14: controller; - 15: image analysis module; - 111: seated; - 112: file; - 113: headrest; - 1001: first operation; - 1002: second operation; - 1003: third operation; - 1004: fourth operation; - 1005: fifth operation.
Claims
Claims
1. Method for assisting in correcting the posture of an occupant of a vehicle seat (11) comprising: a. collecting (1001) as input signals from a set of sensors (12), said sensors being jointly arranged so as to detect the posture of the occupant of the seat; b. for each time step t, calculating (1002) as a function of the collected signals at least two comfort indices Cijt by implementing at least one algorithm, for at least two of the comfort indices Ci>t, the time step t itself further being a calculation parameter, so that the value of the index Ci>t evolves as a function of the posture of the occupant and also as a function of time; c. for each time step t, calculating (1003) an overall comfort index CGjt as a combination of all the comfort indices Ci>t; d. compare (1004) each of the comfort indices Ci>t and the overall comfort index CGjt to one or more reference values; e.generating (1005) at least one output signal to an alert device (13) arranged to emit an alert, said generated signal being dependent on the results of the comparison.
2. Method according to claim 1 in which the calculation of at least one of the comfort indices Ci>t is furthermore a function of at least one parameter specific to the occupant and previously collected from among anthropometric, physiological, pathological parameters, an age and preferences of the occupant.
3. Method according to one of the preceding claims, in which each comfort index Cijt is associated with a part of the occupant's body.
4. On-board vehicle system (1) comprising: - a seat (11) capable of receiving an occupant of the vehicle; - a set of sensors (12) jointly arranged so as to detect the posture of the occupant of the seat and at least part of which are integrated into the seat; - an alert device (13); and - a controller (14) capable of receiving signals from the set of sensors as input and of generating at least one signal as output for the alert device, said controller being further configured to: a. collecting (1001) as input signals from a set of sensors (12), said sensors being jointly arranged so as to detect the posture of the occupant of the seat; b. for each time step t, calculate (1002) based on the collected signals at least two comfort indices Ci>t by implementing at least one algorithm, for at least two of the comfort indices Ci>t, the time step t being itself a calculation parameter, so that the value of the index Ci>t evolves as a function of the occupant's posture and also as a function of time; c. for each time step t, calculate (1003) an overall comfort index CGjt as a combination of all the comfort indices Ci>t; d. compare (1004) each of the comfort indices Cijt and the overall comfort index CG,t to one or more reference values; e. generating (1005) at least one output signal to an alert device (13) arranged to emit an alert, said generated signal being dependent on the results of the comparison.
5. System according to claim 4, wherein the set of sensors includes at least one of the following sensors: - a pressure sensor; - a temperature sensor; - a humidity sensor; - an interdigital sensor such as a capacitive sensor; - a Hall effect sensor; - a video sensor arranged to capture an image of at least part of the occupant of the seat and connected to the controller via an image analysis module (15); - a position and / or orientation sensor of a first part (111, 112, 113) of the seat relative to a second part (111, 112, 113) of the seat; - a position and / or orientation sensor of a part of the seat relative to a part of the vehicle separate from the seat; - a position and / or orientation sensor for a steering wheel, a seat belt attachment or a vehicle mirror; - a sensor integrated into the back of the seat; - a sensor integrated into the seat base.
6. Computer program comprising instructions for implementing implementation of the method according to one of claims 1 to 3 when this program is executed by a processor.
7. Non-transitory recording medium readable by a computer on which is recorded a program for implementing the method according to one of claims 1 to 3 when this program is executed by a processor.
