Loading...

Design and Implementation of Fall Detection System Using Body Movement

Tavakkoli Anbarani, Ali Mohammad | 2015

678 Viewed
  1. Type of Document: M.Sc. Thesis
  2. Language: Farsi
  3. Document No: 48817 (05)
  4. University: Sharif University of Technology
  5. Department: Electrical Engineering
  6. Advisor(s): Vosoughi Vahdat, Bijan
  7. Abstract:
  8. Measurement is the start point of identification and analysis of a physical quantity. By recording body movements, one obtains an efficient insight of musculoskeletal performance. Furthermore it leads to quantification of physic-structural health of human. In addition this information can be applied to many human related applications such as: balance survey, sport analysis, muscular pathology, physiology method performance evaluation, etc.With reference to statistics of the elderly fall, the large number of medications with bilateral effects such as instability and drowsiness, arthritis resulted from daily movement misconduct, there is a need for a preemptive system to overcome these injuries.Previous researches suggest body movement recording mechanisms that are terrain-specific and non-portable which to some extent, hinder their functionality in daily use applications.In the current research the purposed system is designed and manufactured to simultaneously measure and record body movement while being portable and with no interruption to user. Using the data recorded by the purposed system, a method for prevention of elderly falling is introduced.The purposed system is comprised of MEMS inertial sensors including accelerometer, gyroscope and magnetometer. Then the system is applied to two specimen. The recorded data were introduced to a two-layer supervised artificial neural network (ANN) by means of one of feedback propagation method variance. The purposed system was able to categorize the test data with regression factor of 9.81834×10-1 and MSR of 8.04642×10-1
  9. Keywords:
  10. Inertial Sensor ; Artificial Neural Network ; Accelerometery ; Motion Sensor ; Falling Detection

 Digital Object List

 Bookmark

No TOC