Research On Fall Detection System For Solitary Elderly Based On Wearable Equipment

With the increase of the number of the elderly living alone in our country,the safety of the elderly living alone has attracted more and more attention.After falling,the elderly living alone cannot be timely warned and rescued,resulting in serious consequences and even endangering their lives,making falling the first factor threatening the safety of the elderly living alone.Therefore,the design of a wearable system that can accurately identify and actively alarm the elderly when they fall has a more prominent engineering value and social significance for ensuring the safety of life and improving the quality of life of the elderly living alone.Based on the kinematic characteristics of human falls,this thesis designs a fall detection system based on wearable devices.The advanced nine-axis motion sensor is used to acquire real-time information of human pose angle.At the same time,the fall detection algorithm based on acceleration sensor is combined with human pose detection algorithm,and a fall detection algorithm based on multi-threshold is designed to improve the accuracy of fall detection.The main work of this thesis are as follows:1.The practical significance of researching the fall detection system for the elderly living alone is analyzed and elaborated.The existing fall detection technology is introduced.Finally,the fall detection scheme based on wearable equipment and the fall judgment method based on threshold are determined.The theory of human motion posture and fall is introduced.The optimal position of the detection system is determined by the proportion of the mass of each part of the human body.The falling process of human body is analyzed in detail by establishing the Cartesian coordinate system of human body in combination with the optimal position.At the same time,the parameters of human posture angle are introduced to calculate the posture.2.Define system requirements.The overall scheme design and hardware platform design of the system are completed.The selection of microprocessor,human motion sensor,communication positioning module,data storage module and power management module and system circuit design are completed and the program is written and tested in practice.3.Complete the software design of the system.The software focuses on the data acquisition and processing.Besides,the software achieves a multilevel threshold fall detection algorithm with human acceleration,angular velocity and attitude angle as its characteristic parameters.Respectively,test and verification has been conduct on the algorithm of the two fall detection algorithms with "acceleration + angular velocity" and "acceleration + angular velocity + attitude angle" as their characteristic parameters.Experiments show that the accuracy of fall detection can be further improved by introducing a fall detection algorithm with human attitude angle parameters.This thesis realizes the hardware and software design of the wearable fall detection system for the elderly living alone,and studies the optimization and implementation of the fall detection algorithm.The design has a certain practical significance for the relevant engineering applications and theoretical research.