Falling monitoring system based on multi axial accelerometer

A portable & wearable falling monitoring system is proposed in this paper, which is based on microcontroller and triaxial accelerometer, and used for real-time monitoring of human body motion state. Its purpose is to make real-time judgments during the falling, and enable airbags or other protective devices before the body impacts with the ground, preventing or reducing the damage of falling to the human body. This paper proposes a multi-condition comprehensive evaluation algorithm based on the value of acceleration SVM, by analyzing acceleration curve of the processes of the body falling and various daily activities. In the algorithm, the weightlessness condition, the time condition, the speed condition, and the angle condition etc are considered in human body motion state analysis, then the judgment is made whether a falling is happening by these conditions. Experiments show that the method can effectively identify the course of the falling with a low miss report rate and false report rate, meanwhile it can provides the appropriate time margin, and should be a feasible and practical way.