AU2017101323B4 - LifeChair, A system which tracks a user’s sitting posture and provides haptic feedback through a pressure sensory chair or chair cushion to encourage upright posture. - Google Patents

Abstract

LifeChair - Abstract A system which tracks a user's sitting posture and provides haptic feedback through a pressure sensory chair or chair cushion to encourage upright posture while sitting. The sensory component of the seat utilizes a predetermined array of pressure sensitive sensors that detect both the presence and posture of the user. The sensor array can calibrate to the user's sitting conditions using a guided setup routine implemented in an application. The algorithm for detecting states of sitting posture is applied as a classification table. When the system detects incorrect sitting posture such as slouching, haptic feedback will be directly delivered to the user through vibration or sound, preferably through the chair. Furthermore, the vibration location on the user's back is relative to the area in need of immediate posture correction. A developed haptic feedback language is used to communicate to the user different conditions which require attention. Depending on the frequency, magnitude and pattern of the vibration, the user will determine if the system is alarming them to correct posture, stand up and take a break or to drink water.

Description

LifeChair - Description

Title of Invention

LifeChair

Technical Field [0001] The technical field of this invention crosses between ergonomics, health care and artificial intelligence. It relates to a chair which reminds its user to sit with good posture and to regulate their sitting time.

Background [0002] In the modern world, millions of people worldwide are spending an ever-increasing amount of time sitting down. In particular, office worker health is a growing problem for people who sit in office chairs for extended periods of time. When sitting down there is a tendency to sit with poor posture. Poor sitting postures, over time, may cause a variety of negative health effects that are both mental and physical. This has also turned into a growing problem for employers, who have to shoulder the economic burden of these health effects.

[0003] These adverse health effects can include weight gain, increased risk of diabetes, increased risk of heart disease, deep vein thrombosis, shoulder-, back- and neck-pain, heightened stress, and an increased tendency to headaches. Compression of the core also causes poor digestion and a decreased oxygen intake. This can lead to gastrointestinal issues, a slowed metabolism and a decreased oxygen flow to the brain. Over a long work day this can result in increased stress levels and a decrease in productivity. Studies have also linked extended seating time to increased risk of psychological distress, depression and irritability.

[0004] Further, incorrect sitting posture can cause musculoskeletal ill-effects such as spinal imbalance, back problems, sciatica and decreased range of motion in the hip joint. Many people lean forward using their neck when looking at computers in order to counteract poor posture, which strains the muscles in the shoulders and neck.

Summary of Invention Technical Problem [0005] An ordinary chair does not require its user to use it correctly. One of the hardest obstacles to good posture is that the onus is on the user to sit correctly, on an almost constant basis. Ergonomic chairs are available, but these are not widespread due to lack of portability, high cost and reliance on the user to regulate their own sitting behaviour. Further, these devices are passive in nature, and do not notify the user if they are using the device correctly and sitting with proper posture. There are other health-care focused seating, such as massage chairs, but these are currently focused on treating existing conditions rather than prevention.

[0006] Our invention has been specifically designed to take the ubiquitous chair, add an element of artificial intelligence and train the user to sit correctly with upright posture through a variety of feedback loops.

Solution to Problem [0007] The LifeChair in accordance with this invention comprises an array of pressure sensor pads, embedded in either a seating or back cushion, whether built into a chair or as an attachment, an algorithm which analyses the sensor pads, a control module, power supply, feedback mechanisms which may include vibrations and/or sound, and an accompanying computer application.

[0008] The sensor array is a set of electronic pads which respond when pressure is applied. They must be constructed to withstand the force applied when a human sits. The preferred method of use is fabric sensors with conductive sheet layer to act as the conductor. This is the first use of such sensors to create a pressure sensing design.

[0009] The product when in use according to this specification is able to analyse the information from the sensor array using a computer algorithm to determine the sitting position of a person. The best practice of classification of sitting posture is done based on a calibrated frame of reference of the user’s body shape and chair. This calibration is done during setup of the LifeChair to create a personalized solution. Alternatively, there are default posture templates included in LifeChair which may be used.

[0010] A control module also exists to run the electronic components, which consists of a microprocessor, battery, and integrated power module. The battery is rechargeable via a micro USB charging port. The LifeChair is preferably able to remain operational for over 30 hours of continuous use until it needs to be recharged, in order to. The control module also consists of a low energy Bluetooth module for communication between the microprocessor and the mobile device for the purpose of transmitting sensor data and other information to the device running the LifeChair application.

[0011] The LifeChair also preferably comes with a computer application which allows for processing and control of mechanisms in the device. This application can also act as the feedback mechanism. The connection from the control module to a computer device is preferably done wirelessly, most preferably via Bluetooth, but can also be wired. The LifeChair app contains several features which can include but are not limited to posture tracking and training through classification and vibrotactile feedback, Stretching classification and reminders, standing break reminders, meditation classification, guidance and reminders, water drinking reminders, customization of strictness of posture alerts, a method for visualizing and tracking posture and activity statistics and improvements, a collection of algorithms that learns from all other LifeChair devices connected to the network to improve upon itself, a module which allows the user to create their own custom reminders, and a module that synchronizes to the calendar schedule of the user to remind them of events or meetings.

[0012] If built as an attachment, LifeChair preferably utilises a set of straps to tie to a chair it is placed on. These straps and the associated buckle can contain the control module and/or electronic components. The preferred form of the straps is a double strap design with an adjustment mechanism at the backside top and bottom of the cushion. A single adjustable strap for the middle of the cushion can be preferred for cushions designed for smaller chairs.

Advantageous effects of Invention [0013] The LifeChair is superior since it takes the self-regulating function out of the user of a chair and replaces it with advanced artificial intelligence. This occurs through the analysis of the sensor pads and the feedback provided to the user when the algorithm detects that the user is not sitting upright. The feedback loops provide the user with an automatic reminder system. We acknowledge there is still an element of self-regulation, because the user still has to use the LifeChair and then physically action the feedback received from the device, but there is a strong self-selection and motivation factor attached.

[0014] The integration of artificial intelligence into a chair is another key advantage. The functions of the app include but are not limited to showing the user feedback on seating performance, providing a medium for feedback and provide a function for peripheral areas such as regulation of seating time and monitoring of hydration. Furthermore, the app allows the user to pre-set the standing reminders of LifeChair based on their working conditions. Although we offer recommended time intervals, we also allow customization on the user’s personal needs. The user may also set the frequency at which their posture is tracked, this can be done in real time or at set intervals in order to conserve battery life.

[0015] Potential applications include (but not limited to) home, the workplace, motor vehicles, or airplanes. The product is targeted at people who spend a majority of their day setting down or in an otherwise sedentary position, such as office workers. The device is also targeted at people who have a history of back or musculo-skeletal issues, including for use as a rehabilitation aid.

[0016] It is intended that an ordinary user would use the LifeChair to monitor their posture while they are sitting down. Once a user intends to sit for a period of time, they would turn on the unit and have the smart phone application installed. If the user is using the product for the first time then a calibration process will commence. During the normal course of operation the product will alert the user through feedback loops when their posture has deteriorated, or when they have been sitting for too long and encourage them to take a break. We have developed a specialized haptic feedback language to help users distinguish between the different alerts of LifeChair. Depending on the frequency, magnitude and pattern of the vibration motors, the LifeChair could be alerting the user to either correct their posture at a set location, stand up and take a break from sitting or to consume water to remain hydrated.

Brief Description of Figures

Figure 1 is an illustration showing the body of the product.

Figure 2 is an illustration of a user with LifeChair installed on their chair sitting in both a forward slouching position (left) and a correct upright position (right).

Figure 3 is an illustration of some of the postures which can be detected by the product, in this case leaning to the left and right.

Figure 4 is an exploded view of the interior of the LifeChair. It shows the layers of the product, from front to back including the top cover, front layer of sensor pads, middle cushioning layer, back layer of sensor pads and back cover. The second and fourth layer in the drawing are made from conductive fabrics. The middle layer is made from velostat or similar resistive material.

Figure 5 is a schematic illustration of the front sensor layer outlined in Figure 4. The sensor array is an exemplary pattern with some exemplary dimensions shown. This front layer acts as the main conductive layer and works in conjunction with the back layer outlined in Figure 6.

Figure 6 is a schematic illustration of the back sensor layer outlined in Figure 4. The sensor array is an exemplary pattern with some exemplary dimensions shown. This back layer acts as a receptor to the front layer shown in Figure 5 and is designed to mirror the front sensor layer.

Figure 7 identifies a possible arrangement of the vibration motors within the LifeChair cushion, denoted by letters A to D.

Figure 8 is an exploded view of a possible 5 layer cushion design. The layers from front of cushion to back are front cushion, sensor layer, conductive layer, receiving sensing layer and back cushion.

Figure 9 is an oblique view of Figure 8, illustrating how the layers of the product line up.

Figure 10 is a collapsed view of Figure 9, illustrating how the layers of the product are compiled from front to back.

Figure 11 is the schematic of data flow through the entire product between the device, app and server.

Figure 12 is the schematic of data flow through the seating regulation feature. Based on user inputted settings, the app will track seating time and send information to the chair when it is time for the user to stand up.

Figure 13 illustrates various postures which can be experienced while using the product, including left slouching, right slouching, front slouching, front slouching side view, front slouching severely, limited contact, no lower back contact, no contact, and upright.

Figure 14 illustrates the locations of various vibrational feedback mechanisms corresponding to the postures shown in Figure 13. No feedback is provided for No Contact and Upright postures.

Figure 15 is a simple schematic illustration of the LifeChair without the sensor array.

Figure 16 is a schematic illustration of the LifeChair with all measurements included. This image forms part of the design specifications for manufacturing.

Figure 17 is a schematic illustration similar to Figure 16, but showing alternative measurements. This image forms part of the design specifications for manufacturing.

Figure 18 is a drawing of the back of the LifeChair utilising a one strap design. Point 1 is where the strap attaches to the chair; point 2 is the buckle which the stat fits into; point 3 represents the body of the chair; and point 4 is the strap.

Figure 19 is a design schematic identifying the location and position of the logo on the front of the chair and strap.

Figure 20 is a drawing showing the side profile of the LifeChair, and how it has been designed to curve based on the curvature of the lumbar.

Figure 21 are various drawings of the LifeChair, showing the difference between the single strap and double strap version.

Figure 22 is a drawing of the control module box which is attached to the side of the chair as can be seen in figure 1. The top hole is space for an LED or similar light. The middle hole is designed to fit a power button for turning the device on and off. The lower elongated hole is designed as a charging port, preferably USB or micro USB.

Figure 23 is an illustration showing the location and function of the zipper at the bottom of the cushion unit which allows access to the inner components, including the sensor array and foam layer. The zipper is located on the bottom edge of the cushion and opens towards the control module on the right-hand side.

Figure 24 is a schematic diagram showing the location and dimensions of the vibration motors when installed into the cushion. The areas identified as A, B, C and D are individual motors with corresponding circuitry going back to the control module.

Figure 25 is an exploded view of the pressure sensing system. The layers from left to right are front cushion, sensor layer, conductive layer, receiving sensing layer and back cushion.

Description of Embodiments [0017] The body of the LifeChair is designed to mimic the shape of a chair. It can be designed either as an overlay to an existing chair, a complete chair with the system embedded or as a standalone unit to be attached to a chair. If being manufactured as an overlay, the main section of LifeChair mimics the back of a chair, with the shape being approximately rectangular to match the back of an existing chair, as seen in figure 1. The edges may be tapered in order to provide compatibility with a wide variety of chairs. The bottom or seat component, that is placed under the user is optional if the unit is built as an overlay. Furthermore the LifeChair system can be developed as 2 components linked to each other. In this design the main cushion is behind the user’s back and the secondary cushion is placed under the user’s bottom.

[0018] The body section has optional components which are designed to hold the product in place during normal use. This is intended to counteract the effects of the unit sliding around the chair, which may cause incorrect measurement of posture and impair the intended function of the product. These components preferably take the form of straps which allow the product to be fastened to an existing chair. If the product is built as an entire chair these components are not required.

[0019] The body section is manufactured from materials which mimic those used to manufacture existing furniture. The preferred material used on the front and back is 3D mesh fabric. Alternatively, the LifeChair may also contain polyester, cotton and velcro.

[0020] The body unit contains all of the components required for LifeChair to function, this includes at a minimum the sensor array, battery unit, and the control module. The sensor array is at the forefront of the cushion interior, with the intention that it not be easily felt by the user when sitting on the product during the normal method of operation. The preferred internal structure of the body unit is a soft outer layer which the user will rest on (“the front”), a soft structure to provide comfort support and which contains the sensor array and the vibration motors, preferably being a type of foam (“the middle”), and then a harder outer layer which will rest against an existing chair (‘the back”).

[0021] Additional components may be added to the body of the device in order to increase the ability of the product to perform its intended function, or in order to perform additional functions. The system is designed to be extended in the future for additional functions. One possible use case is the tracking of other human biological signals which could include components, but are not limited to, a heart rate monitor, hydration monitor or blood pressure monitor.

[0022] The pressure sensors are built into the body component outlined above. The sensors themselves comprise of electronic components which react when pressure is applied, in the regular course of operation of the product this pressure will be exerted by the user sitting down on the product while it is attached as an overlay to a chair. The preferred structure or pattern of the sensors is a 9 sensor array where there are 3 rows and 3 columns of sensors.

The pressure sensors may be different in pattern to support different use cases. For example, a denser set of sensors such as 4 rows by 4 columns would increase the device’s ability to map posture. The middle column of sensors are arranged along the centre of the cushion so to align with the user’s spine. The 1st and 3rd columns are symmetrical about the centre column. The sensors are circular in shape. This structure of sensors allows for accurate tracking of the vital posture points of the back. This includes pressure tracking of the shoulders, lumbar, obliques, along the spine and sides of the back.

[0023] The pressure sensors are made of conductive fabric materials and are connected through internal wiring to other components and the control module. The sensors behave as force sensitive resistors, where the amount of pressure detected is relative to the force applied on the sensor. This force is derived from the amount of contact made between the 2 separated elements through a resistive conductive film layer.

[0023a] The pressure sensing technology uses 2 layers of conductive fabric separated by a conductive film, preferably velostat or similar material. An optional thin layer of foam is added in between the top layer of conductive fabric and the conductive film layer to regulate the pressure activation range. The sensors are connected to a PCB circuit, which measures an analogue range of pressures based on the resistivity of the sensor under a given force. The 2 layers of conductive fabric can be denoted as (1) the top layer, which contains the individual isolated pressure sensors, which are each individually connected by a single wire to an input port on the PCB, and (2) the bottom layer, which contains a single sheet or connected pieces of conductive fabric that represent the ground plane. This layer has a single connection to the ground pin on a PCB. The middle layer uses conductive film, velostat or a similar electrically conductive material, this provides a layer of resistivity between the 2 layers of conductive fabric. The resistance changes based on the force applied to the sensor. This force can be interpreted by reading the analogue signal from the individual sensors of the top layer of conductive fabric. An extra layer of foam with holes may be added between the top layer of conductive fabric and the middle layer of conductive film to regulate the force required to activate a pressure sensor. In the case where the foam is placed on top of the middle velostat layer, the foam should have holes through it to allow the conductive layers to still interact through the velostat when enough pressure is applied.

[0024] The algorithm is built as part of the computer application and will determine whether the user is sitting with correct or incorrect posture. The algorithm is personalized to each individual user based on an initial calibration when the chair is used for the first time, which will determine a baseline posture for each user. This baseline posture may vary based on the size, shape or weight of a particular user, and can be measured against a standard posture profile to ensure the baseline posture is in an acceptable range. The algorithm is tuned to detect imbalances in seating posture during the calibration phase to ensure a poor calibration is avoided. The algorithm is able to detect the user in a variety of positions including but not limited to sitting upright, slouching, hunching, slumping, leaning to the side, leaning forward or leaning backwards. Some of these positions are shown in figures 2 and 3.

[0025] The algorithm works as a classification model. Based on the output of the sensors pads, the product can categorise how the user as sits as one of a pre-defined set of postures. In order to detect posture, each sensor collects data at a predefined intervals of time. These values are then compared against a baseline set of calibrated force readings which is personalized to each user in order to determine the deviation from normal posture.

[0026] When the deviation at particular sensors exceeds the threshold, a posture is then predicted based on the classifications provided in Table 1.

Table 1.

[0027] In Ihe above table, there are 11 possible posture states. There are 9 abnormal states, one upright state, and one null stale where there is no user present. The function of deviation is e(t)i where e is the computed deviation at time 1 at sensor i. The state look-up table uses a variable γ to perform a priority based search where it will predict the posture using a top-down look-up approach that ensures the state detected ranges from critical to specific. The sensors and vibrational motors, referred to in Table 1, are referencing their positions in Figure 7.

[0028] The Feedback mechanism is intended to function in a variety of ways. Feedback can be provided to the user Visually (either physically on the product or via electronic means), through audio or through touch. The visual feedback mechanism can be provided through the computer application, through a notification to a smartphone, or through a message in the application. Audio feedback is provided through the application or via a speaker attached to the body of the product. The preferred function is that the user is able to customise the frequency and intensity of the visual and audio feedback loops.

[0029] The physical feedback mechanism uses haptics to reinforce posture-positive behaviours in the user. These can be, but are not limited to, vibrations of a phone running the accompanying computer application, and also through direct application of physical stimuli to the user through the product via vibration motors.

[0030] The vibration motors are placed within the body of the product as outlined above, and illustrated in Figure 7. These motors are placed in strategic locations such that the haptic feedback is associated with a specific area. This may include specific types of vibration responses based on which position the user is deficient in. For example, if the algorithm detects the user is leaning to the left with the right side of the body off the sensors, the user might receive feedback in the form of vibration on the left side, encouraging to correct their position. This haptic feedback may also vary in intensity, with the vibration becoming more intense as the length of time in an incorrect posture position increases.

[0031] The LifeChair system uses a specially designed language of haptics to communicate different messages to the user. The alerts can include but are not limited to Posture

Correction, Standing Reminders or Hydration Reminders.

[0032] Different types of haptic streams are defined by various combinations of vibration identifiers which relate to the location of that disc on the chair, frequency of the vibration and magnitude. For posture correction feedback the haptic cues are normally location based pulses of one or more motors simultaneously. For seating regulation, we use a 'ramp up' effect where 2 or more vibrators simultaneously begin to vibrate continuously. The vibration begins with small strength and then increases to a high strength vibration incrementally. This ramp effect distinguishes itself from the pulsing effect so the user can differentiate what the LifeChair is communication, and then respond appropriately to the feedback.

[0033] Posture Correction haptic feedback is described in Table 1 which outlines the motors that are activated depending on which posture is detected. The motors are pulsed at a default frequency of 2.5Hz at 75% -100% vibration magnitude but these settings can be altered depending on their effectiveness to influence the user's behaviour.

[0034] Standing reminders are triggered at set intervals defined in the app by the user. When the LifeChair system has deemed it is time to take a break and stand up, the vibration motors begin pulsing in a special formation to communicate this to the user. This can be distinguished from posture correction by the increased frequency of the vibration for reminding the user to stand up. By default the frequency of vibration is set to be 5Hz and at full vibration magnitude. The pattern of vibration is pulsed as [A B], [C D], [A B], [C D] (using the motors labelled in figure 7) and so forth until the LifeChair system has detected the user had stood up or disabled the alert through the app.

[0035] Hydration reminders are triggered at intervals defined in the app by the user. When the LifeChair app determines it is time to take a drink of water it will begin activating the vibration motors in the arrangement [A], [B], [D], [C], [A], [B], [D], [C] and so forth until the user has notified the app that they have taken a drink of water. The pulsing is at a default vibration frequency of 1Hz and 50% - 75% magnitude.

[0036] For stretching coaching, we use patterns of pulsing motors to signal the timely reminders to stretch, start of a stretch, completion of an individual stretch and completion of all stretches. Each is a different series of pulses with a different frequency and strength of vibration to that set of posture reminders. This is to help the user distinguish between posture tracking and stretching modes. The pattern of vibration to alert the user to begin stretching is pulsed as A, B, D, C, A, B, D, C, A, B, D, C. During the exercise, the pattern to move to the next stretch is pulsed as [A B] [A B] [C D] [A B]. When all stretches are completed the vibrations will be pulsed as A, C, B, D, A, C, B, D, A, C, B, D to indicate this routine is finished.

[0037] The user is able to modify the vibration frequency and magnitude, but this will affect all of the vibrational settings equally. This results in the patterns and relative frequency between haptic streams remaining equal, enabling the user to distinguish different types of vibrational reminders. The motors are preferably pulsed at 200ms intervals that is, 200ms ON and 200ms OFF. The frequency and power of vibration can be set by the user to optimize for their response.

Claims (5)

1. LifeChair - Claims

   1. LifeChair, which is a system for real time detection and correction of posture, comprising of a chair with a cushion or similar padded surface, straps to hold the cushion or similar padded surface in place if required, an array of conductive fabric pressure sensors built into the cushion or similar padded surface, a control module attached to the sensors which also controls a feedback mechanism, the feedback mechanism built into the chair which is controlled by the control module, a power supply which powers the control module, pressure sensors and feedback mechanism, posture classification algorithm which interprets data from the pressure sensors and classifies a user's posture to at least the extent of identifying it as correct or incorrect, and smart phone application for the visualisation of pressure data derived from the posture classification algorithm and classified posture and customization of system parameters, including at least one of strictness of posture alerts and time between standing break reminders, water drinking reminders or exercise reminders.
2. 2. The system from claim 1, where the array of conductive fabric pressure sensors comprise: A first layer including M isolated conductive fabric sheets that represent M individual sensors and may be trimmed to different shapes and are electronically connected to a circuit via M wires; A second layer including M isolated conductive film sheets, which operates with a variable resistivity in response to applied pressure; A third layer of N conductive fabric sheets that combine to cover the surface area of all components of the first and second layer, it is electronically connected to a circuit and a common ground pin; An optional layer of foam with holes that is placed between the first and second layer and behaves as a physical resistor, wherein the holes in the foam allow the first and second layers to interface when enough force is applied to compress the foam layer, wherein the thickness and density of the foam is varied to adjust force thresholds of the system, where a foam with greater thickness and density will increase a force tolerance of the pressure sensors.
3. 3. The system from claim 1, where the feedback mechanism is haptic in nature, comprises of one or more vibration motors placed at different locations within the cushion or similar padded surface, is activated by the posture classification algorithm ora reminder feature, will activate in different patterns by activating more than one vibration motor simultaneously to communicate cues to the user, inbuilt into the cushion or similar padded surface, for the purpose of instructing the user to correct their posture, perform an exercise, take a standing break or to take another action.
4. 4. The system from claim 1, where the posture classification algorithm is composed of: Data from the pressure sensors as a form of input, where individual pressure sensor signals are processed as a collective data input array; A set of pre-determined postures that have been modelled in this system based on the number of individual conductive fabric sensor pads used; A calibration data array of individual pressure sensor values taken at time t = 0 and is used as a reference point for threshold based calculations. The calibration data array represents a set of values for an ideal posture which is determined by approximating vertical and horizontal symmetry in the pressure sensor values; A threshold algorithm that processes the collective data input array of the individual pressure sensors and determines if a particular sensor at time t is experiencing an error if its signal has surpassed a threshold integer value in reference to the calibration data array; A classified posture based on collective errors derived from the threshold algorithm process.
5. 5. A system as defined in claim 4 where the system can be adjusted via the smartphone application to biomechanics characteristics and preferences of a user via a threshold-based calibration mechanism that performs an approximate vertical and horizontal pressure balance check for the purpose of defining an ideal posture reference point.