AU2021102699A4

AU2021102699A4 - A system and method for smart walking assistant for elderly care

Info

Publication number: AU2021102699A4
Application number: AU2021102699A
Authority: AU
Inventors: Pratik Bhattacharjee; Suparna Biswas
Original assignee: Biswas Suparna Dr
Current assignee: Biswas Suparna Dr
Priority date: 2021-05-19
Filing date: 2021-05-19
Publication date: 2021-11-04
Anticipated expiration: 2029-05-19

Abstract

The present disclosure relates to a system and method for smart walking assistant for elderly care. The system consists of four modules: Fall Detection Module (FDM), Posture Identification Module (PIM), Location Handling Module (LHM), and Integration Module (IM). In this disclosure a micro electro-mechanical system (MEMS) based sensor (MPU6050) is used to collect the live 3 axis accelerometer and 3 axis gyroscope data and the data is used to detect the fall and posture in elderly people and this system is implemented on common walking stick. As soon as fall without self recovery is detected the IM sends a warning message to both monitoring module and SMS to relative/s of the subject with the date, time, and posture before the fall and the location of the fall. The location of the fall is detected using ESP8266 Wi-Fi module and Wi-Fi positioning system (WPS) in LHM. 17 100 Fall detection module (FD M) 102 Posture identification module (PIM) 104 Locat ion handling module (LHM]106 Integ ration module (1M)108 Figure1 200 Taking a raw live 3 aus accelerometer data and using moving average filter to normalize 202 said accelerometer data applying a minimum and maimum threshold value to detect a fall which is confirmed bf A 2 sumn vector magnitude (SVM) based approach rece g live 3 aids accelerometer and 3 ads gyoscope data from MPU6050 and thereafter combining using a complementary filter analying combined data using random forest machine learning approach to 20etec a 8 posture of a sbject against a trading dataset rv acquidrnglocation of a smartwalking assistancefrom ESPS266 Wi-Fimoduleandusing Wi-Fi positioning system (WPS) to send a co-ordinate to cloud for obtaining corresponding address detecting and updating location when said subject moves from one location to another 2012 viaavailable hotspots returmng last stored location in case it does not receive a device data, and timestamp of 214 last location< 10 minutes ago, attaching a special message integrating input from FDM, PIM and LHM and preparing an output warning message if there is a fall or remains silent otherwise In an event of a Fall (Fall without self-recovery, wherein ian 216 event of a fall (Fall without self-recovery) said integration module sends a popup waring to a monitoring station along with an audible alert and a copy of said popup warning is sent to a SMS gateway that passes it on to a phone of relative/s of said subject Figure2

Description

Fall detection module (FD M) 102 Posture identification module (PIM) 104

Locat ion handling module (LHM]106 Integ ration module (1M)108

Figure1

200 Taking a raw live 3 aus accelerometer data and using moving average filter to normalize 202 said accelerometer data

applying a minimum and maimum threshold value to detect a fall whichis confirmed bf A 2 sumn vector magnitude (SVM) based approach

rece g live 3 aids accelerometer and 3 adsgyoscope data from MPU6050 and thereafter combining using a complementary filter

analying combined data using random forest machine learning approach to a 20etec 8 posture of a sbject against a trading dataset rv

acquidrnglocation of a smartwalking assistancefrom ESPS266 Wi-Fimoduleandusing Wi-Fi positioning system (WPS) to send a co-ordinate to cloud for obtaining corresponding address

detecting and updating location when said subject moves from one location to another 2012 viaavailable hotspots

returmng last stored location in case itdoes not receive a device data, and timestamp of 214 last location< 10 minutes ago, attaching a special message

integrating input from FDM, PIM and LHM and preparing an output warning message if there is a fall or remains silent otherwise In an event of aFall (Fall without self-recovery, wherein ian 216 event of a fall (Fall without self-recovery) said integration module sends a popup waring to a monitoring station alongwith an audible alert and a copy of said popup warning is sent to a SMS gateway that passes it on to a phone ofrelative/s of said subject

Figure2

A SYSTEM AND METHOD FOR SMART WALKING ASSISTANT FOR ELDERLY CARE FIELD OF THE INVENTION

The present disclosure relates to a system and method for smart walking assistant for elderly care.

BACKGROUND OF THE INVENTION

Fall of the elderly adults is a serious problem since they may result in substantial injury and may even lead to death. The fall can occur anywhere, it can occur at the indoor, on staircase or outdoor. A financially weak person can't afford a care-giver.

To classify the physical condition of the seniors a frailty index is used. The five pre classified conditions for classification are slow walking speed, low physical activity, weak grip strength, exhaustion, and unintentional weight loss. The aging people who possess three or above among these five conditions then they are at risk.

The determination of posture just before the subject fell down is important; it will help in estimating the damage and will provide a hint whether subject is suffering from a particular disease.

In one existing solution, HONEY (Home healthcare sentinel system), a three step detection scheme was proposed; the system consisted of an accelerometer, audio, image and video clips. The fall was detected by leveraging a triaxial accelerometer, speech recognition and on demand video. The system showed an accuracy of 94%. In another existing solution, a technique based on the walk recognition was proposed. This increases both usability and trustworthiness of a smartphone-based fall detection system. This technique automatically determines the orientation of the device; the orientation is used to infer posture and eliminating the false alarm. Another existing solution proposes a model for efficient fall detection. The model was based on WIFI with the pointer to activities using physical layer channel state information with the 3 indoor scenes arrangement of transmitter and receiver links. To analyzed the performance random forest algorithm was used which involves support vector machine of single classifier with the precision at average rate. In another existing solution a RF- Based fall monitoring system called "Arokee" uses a convolutional neural network and an FMCW radio equipped with vertical and horizontal antenna arrays. It uses spatio-temporal convolutional and abstract complex spatio-temporal patterns. The system uses two CNNs, one for fall event and other one is for stand up event. Another existing solution stimulates the fall detection through the triaxial accelerometer. It monitors the activities of daily living (ADL). It yielded 100

% sensitivity with a number of false alarms. In another existing solution a system based on artificial neural network (ANN) detects a fall using accelerometer data from a wearable sensor mounted on the wrist of the person. In another existing solution a fall detection system based on accelerometer uses SVM and a sliding window to extract 12 point features from the accelerometer data. The fall detection is done using the posture and threshold. The system showed accuracy above 99.9%. In another existing solution, an upper and lower threshold-based approach is used to determine the fall applying Kalman filter, and a 3-DOF accelerometer around the waist of the subject is used. The module has post fall posture recognition with the help of accelerometer angle calculation, and it sends warning if a fall is detected. In another existing solution a meta model is presented for the detecting and prediction of a fall using a HRV analysis on the bases of short term ECG data on hypertensive patients. In this model Naive Bayes was used due to which the model gained the sensitivity, specificity, and accuracy of 72%, 61%, 68% respectively. In another existing solution, the fall detection is based on the posture recognition based on CNN and the fall is detected with the help of laying posture. The data was extracted from a recorded video stream and processed after removing the background. The model showed the accuracy of more than 95%. In another existing solution the horizontal and vertical variation of human posture is used to propose a fall detection system. The SVM is used for the detection process but this system is not suitable for outdoor purpose as it is a camera based system. In another existing solution the data was taken of both normal and sick people and a fall detection technique has been proposed in which KNN based hybrid classification technique has been applied using principle component analysis (PCA) for fall detection and prevention. In another existing solution a RGB camera based technique using Microsoft Kinect technology for automatic detection of instability among Parkinson patients by recording their joint and limb movements. In this technique K- nearest neighbour (kNN), Multinomial Logistic Regression (MLR) and Support Vector Machine (SVM) is used with polynomial kernel on MATLAB for analysis of the posture of the subject. The technique showed 95% accuracy with 2 classifiers and upto 70% with 3 classifiers.

In one prior art solution the apparatus for detection of human falls comprises: an acceleration detector, for detecting vibration events, typically placed on a floor, a microphone, located in association with the acceleration detector for detection of corresponding sound events, and a classification unit to classify concurrent events from the microphone and the acceleration detector, thereby to determine whether a human fall is indicated. If the event appears to be a human fall, then an alarm is raised.

In another prior art solution method and device for fall prevention and detection is presented, especially for elderly care based on digital image analysis using an intelligent optical sensor. The fall detection is divided into two main steps; finding the person on the floor, and examining the way in which the person ended up on the floor. The first step is further divided into algorithms investigating the percentage share of the body on the floor, the inclination of the body and the apparent length of the person. The second step includes algorithms examining the velocity and acceleration of the person. When the first step indicates that the person is on the floor, data for a time period of a few seconds before and after the indication is analyzed in the second step. If this indicates fall, a countdown state is initiated in order to reduce the risk of false alarms, before sending an alarm. The fall prevention is also divided into two main steps; identifying a person entering a bed, and identifying the person leaving the bed to end up standing beside it. The second step is again further divided into algorithms investigating the surface area of on or more objects in an image, the inclination and the apparent length of these objects. When the second step indicates that a person is in an upright condition, a countdown state is initiated in order to allow for the person to return to the bed.

In another prior art solution a fall detection system is proposed which includes a wearable monitoring device that monitors the movement of a person. The device monitors a sensor and detects variation from the normal range and duration thereof. The system determines whether the wearer has fallen through an algorithmic analysis technique using parameters to evaluate the accelerations and timings of the events that comprise a fall. If the combination of the timing and variations from the normal ranges are sufficient as compared to preset thresholds, a fall report will be generated. The wearable device optionally allows qualified professionals to adjust or customize the parameters to optimize the evaluation to the requirements of particular users or classes of users. The wearable device generally transmits data and alerts over a short distance to a console or over a long distance using a connection to a long-distance back haul communication system such as cell network or intemet or both. The device, thus transmit data and alerts to a call center or other designated location.

However, there are different smart home devices available (both as commercial and experimental) those are capable of detecting a fall. However, most of them do not tell any additional information except that a fall has been detected along with time of fall. Therefore in order to avoid the aforementioned drawbacks, there is a need of a system and method for smart walking assistant for elderly care.

SUMMARY OF THE INVENTION

The present disclosure relates to a system and method for smart walking assistant for elderly care. The present disclosure presents a fall and posture detection system by applying machine learning and using low cost sensors. The present work will help especially elder people living unattended. The system can work in both outdoor and indoor. The system can identify fall with and without recovery and the intensity of fall as well. The intelligent live fall with posture detection is designed and implemented exploiting sensors in MPU 6050 (Accelerometer Gyroscope) combined with low cost ESP 8266 processing unit with Wi-Fi connectivity so that the device can be used freely. The system is implemented on common walking stick. In this disclosure the FDS and ADL is used as a common unit for detection and analysis of fall. The system is classified with four classifiers KNN, SVM, Decision tree and random forest; from all these classifiers the random forest gave the best performance. In the case of fall without recovery the alert message with the date, time and location of fall is sent to relatives or caregiver. All the data is stored in a file for future reference. The tests of the system showed the overall detection accuracy of 98%.

The present disclosure seeks to provide a system for smart walking assistant for elderly care. The system comprises: a fall detection module (FDM)to take a raw live 3 axis accelerometer data and uses moving average filter to normalize said accelerometer data and thereby applies a minimum and maximum threshold value to detect a fall; a posture identification module (PIM) to acquire live 3 axis accelerometer and 3 axis gyroscope data from MPU6050 and combines them using a complementary filter and thereafter combined data is analyzed using random forest machine learning approach to detect posture of a subject against a training dataset; a location handling module (LHM) to receive location of a smart walking assistance from ESP8266 Wi-Fi module and using Wi-Fi positioning system (WPS) to send co-ordinates to cloud for obtaining corresponding address, wherein said location handling module automatically detects and update location when subject moves from one location to another via available hotspots; and an integration module (IM) configured with a computing unit to integrate input from FDM, PIM and LHM and prepares an output warning message if there is a fall or remains silent otherwise, wherein in an event of a fall (Fall without self-recovery) said integration module sends a popup warning to a monitoring station along with an audible alert and a copy of said popup warning is sent to a SMS gateway that passes it on to a phone of relative/s of said subject.

The present disclosure also seeks to provide a method for smart walking assistant for elderly care. The method comprises: Taking a raw live 3 axis accelerometer data and using moving average filter to normalize said accelerometer data; applying a minimum and maximum threshold value to detect a fall which is confirmed by sum vector magnitude (SVM) based approach; receiving live 3 axis accelerometer and 3 axis gyroscope data from MPU6050 and thereafter combining using a complementary filter; analyzing combined data using random forest machine learning approach to detect a posture of a subject against a training dataset; acquiring location of a smart walking assistance from ESP8266 Wi-Fi module and using Wi-Fi positioning system (WPS) to send a co-ordinate to cloud for obtaining corresponding address; detecting and updating location when said subject moves from one location to another via available hotspots; returning last stored location in case it does not receive a device data, and timestamp of last location < 10 minutes ago, attaching a special message; and integrating input from FDM, PIM and LHM and preparing an output warning message if there is a fall or remains silent otherwise. In an event of a Fall (Fall without self-recovery, wherein in an event of a fall (Fall without self recovery) said integration module sends a popup warning to a monitoring station along with an audible alert and a copy of said popup warning is sent to a SMS gateway that passes it on to a phone of relative/s of said subject.

An objective of the present disclosure is to provide a system and method for smart walking assistant for elderly care.

Another objective of the present disclosure is to use FDS (Fall Detection System) and ADL (Activity of Daily Living) for detection and analysis of a fall.

Another object of the present disclosure is to give audio/visual warning in the monitoring terminal and send message warning with date, time and location, to the relatives of the subject, whenever a fall is detected for faster emergency help.

Another object of the present disclosure is to identify a fall even when the device become ineffectual after the incident.

Another object of the present disclosure is to minimize the number of warning which is done by considering only without recovery fall for alerts.

Another object of the present disclosure is to store the data for clinical purpose in future.

Yet, another object of the present disclosure is to cross validate across four different classifiers, which are KNN, SVM, Decision Tree and Random Forest.

To further clarify advantages and features of the present disclosure, a more particular description of the invention will be rendered by reference to specific embodiments thereof, which is illustrated in the appended drawings. It is appreciated that these drawings depict only typical embodiments of the invention and are therefore not to be considered limiting of its scope. The invention will be described and explained with additional specificity and detail with the accompanying drawings.

BRIEF DESCRIPTION OF FIGURES

These and other features, aspects, and advantages of the present disclosure will become better understood when the following detailed description is read with reference to the accompanying drawings in which like characters represent like parts throughout the drawings, wherein:

Figure 1 illustrates a block diagram of a system for smart walking assistant for elderly care in accordance with an embodiment of the present disclosure;

Figure 2 illustrates a flow chart of a method for smart walking assistant for elderly care in accordance with an embodiment of the present disclosure;

Figure 3 illustrates the proposed architecture in accordance with an embodiment of the present disclosure;

Figure 4 illustrates the activity wise performance comparison in accordance with an embodiment of the present disclosure;

Further, skilled artisans will appreciate that elements in the drawings are illustrated for simplicity and may not have been necessarily been drawn to scale. For example, the flow charts illustrate the method in terms of the most prominent steps involved to help to improve understanding of aspects of the present disclosure. Furthermore, in terms of the construction of the device, one or more components of the device may have been represented in the drawings by conventional symbols, and the drawings may show only those specific details that are pertinent to understanding the embodiments of the present disclosure so as not to obscure the drawings with details that will be readily apparent to those of ordinary skill in the art having benefit of the description herein.

DETAILED DESCRIPTION

For the purpose of promoting an understanding of the principles of the invention, reference will now be made to the embodiment illustrated in the drawings and specific language will be used to describe the same. It will nevertheless be understood that no limitation of the scope of the invention is thereby intended, such alterations and further modifications in the illustrated system, and such further applications of the principles of the invention as illustrated therein being contemplated as would normally occur to one skilled in the art to which the invention relates.

It will be understood by those skilled in the art that the foregoing general description and the following detailed description are exemplary and explanatory of the invention and are not intended to be restrictive thereof.

Reference throughout this specification to "an aspect", "another aspect" or similar language means that a particular feature, structure, or characteristic described in connection with the embodiment is included in at least one embodiment of the present disclosure. Thus, appearances of the phrase "in an embodiment", "in another embodiment" and similar language throughout this specification may, but do not necessarily, all refer to the same embodiment.

The terms "comprises", "comprising", or any other variations thereof, are intended to cover a non-exclusive inclusion, such that a process or method that comprises a list of steps does not include only those steps but may include other steps not expressly listed or inherent to such process or method. Similarly, one or more devices or sub-systems or elements or structures or components proceeded by "comprises...a" does not, without more constraints, preclude the existence of other devices or other sub-systems or other elements or other structures or other components or additional devices or additional sub-systems or additional elements or additional structures or additional components.

Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs. The system, methods, and examples provided herein are illustrative only and not intended to be limiting.

Embodiments of the present disclosure will be described below in detail with reference to the accompanying drawings.

Figure 1 illustrates a block diagram of a system for smart walking assistant for elderly care in accordance with an embodiment of the present disclosure. The system 100 includes a fall detection module (FDM) 102 which takes raw live 3 axis accelerometer data and uses moving average filter to normalize said accelerometer data and thereby applies a minimum and maximum threshold value to detect a fall.

In an embodiment, a posture identification module (PIM) 104 is used to acquire live 3 axis accelerometer and 3 axis gyroscope data from MPU6050 and combines them using a complementary filter and thereafter combined data is analyzed using random forest machine learning approach to detect posture of a subject against a training dataset.

In an embodiment, a location handling module (LHM) 106 is used to receive location of a smart walking assistance from ESP8266 Wi-Fi module and using Wi-Fi positioning system (WPS) to send co-ordinates to cloud for obtaining corresponding address, wherein said location handling module automatically detects and update location when subject moves from one location to another via available hotspots.

In an embodiment, an integration module (IM) 108 is configured with a computing unit to integrate input from FDM, PIM and LHM and prepares an output warning message if there is a fall or remains silent otherwise, wherein in an event of a fall (Fall without self-recovery) said integration module sends a popup warning to a monitoring station along with an audible alert and a copy of said popup warning is sent to a SMS gateway that passes it on to a phone of relative/s of said subject.

Figure 2 illustrates a flow chart of a method for smart walking assistant for elderly care in accordance with an embodiment of the present disclosure. At step 202 the method 200 includes, Taking a raw live 3 axis accelerometer data and using moving average filter to normalize said accelerometer data; this is done with the help of fall detection module.

At step 204 the method 200 includes, applying a minimum and maximum threshold value to detect a fall which is confirmed by sum vector magnitude (SVM) based approach.

At step 206 the method 200 includes, receiving live 3 axis accelerometer and 3 axis gyroscope data from MPU6050 and thereafter combining using a complementary filter.

At step 208 the method 200 includes, analyzing combined data using random forest machine learning approach to detect a posture of a subject against a training dataset.

At step 210 the method 200 includes, acquiring location of a smart walking assistance from ESP8266 Wi-Fi module and using Wi-Fi positioning system (WPS) to send a co-ordinate to cloud for obtaining corresponding address.

At step 212 the method 200 includes, detecting and updating location when said subject moves from one location to another via available hotspots.

At step 214 the method 200 includes, returning last stored location in case it does not receive a device data, and timestamp of last location < 10 minutes ago, attaching a special message.

At step 216 the method 200 includes integrating input from FDM, PIM and LHM and preparing an output warning message if there is a fall or remains silent otherwise. In an event of a Fall (Fall without self-recovery, wherein in an event of a fall (Fall without self-recovery) said integration module sends a popup warning to a monitoring station along with an audible alert and a copy of said popup warning is sent to a SMS gateway that passes it on to a phone of relative/s of said subject.

Figure 3 illustrates the proposed architecture in accordance with an embodiment of the present disclosure. The architecture of the system contain four modules: Fall detection module (FDM), Posture identification module (PDM), Location handling module (LHM), and Integration module (IM). The brief working of each module is explained below:

Fall Detection Module (FDM): It takes the raw live 3 axis accelerometer data and uses mobbing average filter to normalize the accelerometer data, then to detect the fall the minimum and maximum threshold has been applied. The approach of sum vector magnitude is used to confirm the fall and the module ignores the self recovery falls and generates warning only when the subject fails to recover by himself.

Posture Identification Module (PIM): The module takes the live 3 axis accelerometer and 3 axis gyroscope data from MPU6050 and then the data is combined using the complementary filter, then the combined data is analyzed to detect the posture of the subject using machine learning algorithm against the training dataset of 60,000 data. The training set includes the data of mixed age volunteer of Indian male and female of age groups (10-15 years), (20-30 years), (35-45 years) , (50 - 60 years) and (60- 70 years), the volunteer performs four activities like Walking, Standing, Sitting, and Laying. Based on the live stream the posture is refreshed in every 20 seconds. To identify the posture four classifiers are used KNN, SVM, Random Forest, Decision Tree and it was found that the random forest performs better than any other classifier. The identifies posture is then sent as a output.

Location Handling Module (LHM): This module takes the location of the SMA from the ESP8266 Wi-Fi module and using the Wi-Fi positioning it send the coordinates to the google cloud for obtaining the correct address. When the subject moves the location is automatically detected and updated via available hotspots. The last location of the subject is stored in a file whenever a new location is detected, so that the entire movement of the subject can be tracked, if required. The sensing frequency of this module is of 2 minute in normal mode and an immediate sensing in event triggered mode which means whenever a fall is detected. The module returns the last location in case it doesn't receive any data, and timestamp of last location <10 minute ago with a special message.

Integration Module (IM): This is the final module and it integrates the inputs from FDM, PIM, and LHM and prepares the output warning message, if there is a fall detected and if not the module will stay silent. When a fall is detected which should be a fall without self recovery, the IM sends a popup warning to the monitoring module with the data, time, the last posture before the fall, and the fall location. The same message is also sent to the SMS gateway that sends it on to the phone of the relative/s of the subject. To accelerate the emergency help, group of messaging option can be used to send the message to multiple recipients.

Figure 4 illustrates the activity wise performance comparison in accordance with an embodiment of the present disclosure. The model is tested using four different classifiers Decision Tree, SVM, Random Forest, KNN (K=3). The size of the test data is 1305 and consist four activities- Laying, Sitting, Standing, and Walking. It turn out that the random forest classifier outperforms all other classifiers with accuracy of 98%.

The drawings and the forgoing description give examples of embodiments. Those skilled in the art will appreciate that one or more of the described elements may well be combined into a single functional element. Alternatively, certain elements may be split into multiple functional elements. Elements from one embodiment may be added to another embodiment. For example, orders of processes described herein may be changed and are not limited to the manner described herein. Moreover, the actions of any flow diagram need not be implemented in the order shown; nor do all of the acts necessarily need to be performed. Also, those acts that are not dependent on other acts may be performed in parallel with the other acts. The scope of embodiments is by no means limited by these specific examples. Numerous variations, whether explicitly given in the specification or not, such as differences in structure, dimension, and use of material, are possible. The scope of embodiments is at least as broad as given by the following claims.

Benefits, other advantages, and solutions to problems have been described above with regard to specific embodiments. However, the benefits, advantages, solutions to problems, and any component(s) that may cause any benefit, advantage, or solution to occur or become more pronounced are not to be construed as a critical, required, or essential feature or component of any or all the claims.

Claims

WE CLAIM

1. A system for smart walking assistant for elderly care, said system comprises:

a fall detection module (FDM) to take a raw live 3 axis accelerometer data and uses moving average filter to normalize said accelerometer data and thereby applies a minimum and maximum threshold value to detect a fall; a posture identification module (PIM) to acquire live 3 axis accelerometer and 3 axis gyroscope data from MPU6050 and combines them using a complementary filter and thereafter combined data is analyzed using random forest machine learning approach to detect posture of a subject against a training dataset; a location handling module (LHM) to receive location of a smart walking assistance from ESP8266 Wi-Fi module and using Wi-Fi positioning system (WPS) to send co-ordinates to cloud for obtaining corresponding address, wherein said location handling module automatically detects and update location when subject moves from one location to another via available hotspots; and an integration module (IM) configured with a computing unit to integrate input from FDM, PIM and LHM and prepares an output warning message if there is a fall or remains silent otherwise, wherein in an event of a fall (Fall without self-recovery) said integration module sends a popup warning to a monitoring station along with an audible alert and a copy of said popup warning is sent to a SMS gateway that passes it on to a phone of relative/s of said subject.

2. The system as claimed in claim 1, wherein an approach based on sum vector magnitude (SVM) is configured with said fall detection module to confirm fall of a person upon falling a supporting object.

3. The system as claimed in claim 1, wherein said fall detection module is capable of ignoring fall alike cases (Stumble, Fast sitting, Sudden increase in walking speed etc.) by analyzing duration and threshold values, wherein said fall detection module produces warning only if a subject fails to recover by herself, wherein no warning is generated for average daily activities (ADL), Fell but self-recovered and Fall alike events.

4. The system as claimed in claim 1, wherein said training dataset of is based on a mixed aged volunteer data of Indian male and females from age group of (10-15 years), (20-30 years), (35-45 years) , (50 - 60 years) and (60- 70 years) each performing four activities including walking, standing, sitting and laying, wherein said posture is refreshed in every 20 seconds based on a live stream.

5. The system as claimed in claim 1, wherein last location obtained by said location handling module is stored in a file whenever a new location is detected, wherein later it is possible to track entire movement of said subject for a particular period, if required.

6. The system as claimed in claim 1, wherein said location handling module has a sensing frequency of 2 minutes in general (normal mode) and an immediate sensing in case of a fall, irrespective of time (event triggered mode), wherein said location handling module returns last stored location in case it does not receive a device data, and timestamp of last location < 10 minutes ago, attaching a special message.

7. The system as claimed in claim 1, wherein said warning message consists of a date-time stamp, last posture before fall and a fall location, wherein a group messaging option is used to send message to multiple recipients to accelerate an emergency help for said subject.

8. The system as claimed in claim 1, wherein working steps of said system for smart walking assistant for elderly care comprises:

accompanying a subject under monitoring with a smart walking assistant duly connected to home Wi-Fi through an internet connection; carrying said smart walking assistance by said subject; wherein in case if subject is not able to recover after fall but someone else picks up said smart walking assistant causing change in accelerometer data which is falsely be identified as fall with recovery, is not considered; wherein after a fall if no data comes to said local server for a stipulated period of time, then it is recognized that device becomes unresponsive after fall with assumption that data disruption is caused only with impact of fall not by poor or no network connectivity; wherein said subject takes a sit keeping said smart walking assistance aside and if said smart walking assistance accidentally falls down without subject generating false warning is not considered; refreshing posture and GPS coordinates periodically; and storing data in JSON format which is readily used by a cloud-based system.

9. The system as claimed in claim 1, wherein said smart walking assistance comprises:

a micro electro-mechanical system (MEMS) based sensor (MPU6050) associated with a stick body consists of a 3-axis accelerometer detecting acceleration of said stick body and a 3 axis gyroscope for detecting for measuring orientation and angular velocity of said stick body; a control unit configured with Wi-Fi module for receiving live 3 axis accelerometer and 3 axis gyroscope data from MPU6050 and thereafter combining using a complementary filter and sending said combined data to said cloud through said Wi-Fi; and a battery for providing electrical energy to said control unit.

10. A method for smart walking assistant for elderly care, said method comprises:

Taking a raw live 3 axis accelerometer data and using moving average filter to normalize said accelerometer data; applying a minimum and maximum threshold value to detect a fall which is confirmed by sum vector magnitude (SVM) based approach; receiving live 3 axis accelerometer and 3 axis gyroscope data from MPU6050 and thereafter combining using a complementary filter; analyzing combined data using random forest machine learning approach to detect a posture of a subject against a training dataset; acquiring location of a smart walking assistance from ESP8266 Wi-Fi module and using Wi-Fi positioning system (WPS) to send a co-ordinate to cloud for obtaining corresponding address; detecting and updating location when said subject moves from one location to another via available hotspots; returning last stored location in case it does not receive a device data, and timestamp of last location < 10 minutes ago, attaching a special message; and integrating input from FDM, PIM and LHM and preparing an output warning message if there is a fall or remains silent otherwise. In an event of a Fall (Fall without self-recovery, wherein in an event of a fall (Fall without self-recovery) said integration module sends a popup warning to a monitoring station along with an audible alert and a copy of said popup warning is sent to a SMS gateway that passes it on to a phone of relative/s of said subject.