Modelling human activity through structural vibrations with alternate computational devices to increase cost efficiency

School Name

Spring Valley High School

Grade Level

12th Grade

Presentation Topic

Computer Science

Presentation Type

Mentored

Mentor

Mentor: Juan Caicedo, USC

Oral Presentation Award

1st Place

Written Paper Award

1st Place

Abstract

Every event that occurs has a reaction, whether it be a pebble causing ripples in a pond or a bullet distressing a wall. Within a structure, these vibrations caused by a specific event in a medium can be measured with an accelerometer, and just as the vibrations caused by a bullet observably differ from those caused by a pebble, vibrations caused by walking vary from those caused by falling, running or jumping. To the eye, these differences are slight to severe, but when that signal is dissected, it is identifiably unique by its cause and location with extensive applications from home security to behavior analysis for medical care (including fall detection) to commercial analysis of foot traffic. The focus of this study was to investigate how this signal is collected -- specifically, if a cheaper (and independent) computer could replace a setup that currently costs thousands. The Raspberry Pi was used with an ADXL345 accelerometer as this alternate system. The study included notes of development of the hardware and software as well as analysis of the developed system by comparison to the accepted system. The new system is enabled to continuously read the accelerometer’s z axis output value, maintaining a buffer and saving significant signals. These hypothesized capabilities were confirmed by collecting vibration data from the same impact and comparing how each system recorded the event.

Location

Wall 119

Start Date

3-25-2017 11:30 AM

Presentation Format

Oral and Written

Group Project

No

COinS
 
Mar 25th, 11:30 AM

Modelling human activity through structural vibrations with alternate computational devices to increase cost efficiency

Wall 119

Every event that occurs has a reaction, whether it be a pebble causing ripples in a pond or a bullet distressing a wall. Within a structure, these vibrations caused by a specific event in a medium can be measured with an accelerometer, and just as the vibrations caused by a bullet observably differ from those caused by a pebble, vibrations caused by walking vary from those caused by falling, running or jumping. To the eye, these differences are slight to severe, but when that signal is dissected, it is identifiably unique by its cause and location with extensive applications from home security to behavior analysis for medical care (including fall detection) to commercial analysis of foot traffic. The focus of this study was to investigate how this signal is collected -- specifically, if a cheaper (and independent) computer could replace a setup that currently costs thousands. The Raspberry Pi was used with an ADXL345 accelerometer as this alternate system. The study included notes of development of the hardware and software as well as analysis of the developed system by comparison to the accepted system. The new system is enabled to continuously read the accelerometer’s z axis output value, maintaining a buffer and saving significant signals. These hypothesized capabilities were confirmed by collecting vibration data from the same impact and comparing how each system recorded the event.