Vaulting Detection with the Multi-Model Unscented Kalman Filter

Jesse P. Macht, Josh A. Taylor, Fae Azhari

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

A method for automatically detecting vaulting gait is presented. Two gait models are parameterized based on data from humans, representing a normal gait and a vaulting gait. Structural observability analysis is performed taking into account measurements from a wearable sensor package. A multi-model Unscented Kalman Filter is applied to sensor data from live trials to automatically detect the gait mode.

Original languageEnglish (US)
Title of host publication2024 IEEE 20th International Conference on Automation Science and Engineering, CASE 2024
PublisherIEEE Computer Society
Pages3306-3311
Number of pages6
ISBN (Electronic)9798350358513
DOIs
StatePublished - 2024
Event20th IEEE International Conference on Automation Science and Engineering, CASE 2024 - Bari, Italy
Duration: Aug 28 2024Sep 1 2024

Publication series

NameIEEE International Conference on Automation Science and Engineering
ISSN (Print)2161-8070
ISSN (Electronic)2161-8089

Conference

Conference20th IEEE International Conference on Automation Science and Engineering, CASE 2024
Country/TerritoryItaly
CityBari
Period8/28/249/1/24

All Science Journal Classification (ASJC) codes

  • Control and Systems Engineering
  • Electrical and Electronic Engineering

Fingerprint

Dive into the research topics of 'Vaulting Detection with the Multi-Model Unscented Kalman Filter'. Together they form a unique fingerprint.

Cite this