Related Experiment Video
Updated: Jun 6, 2026
11:16
Engineering Platform and Experimental Protocol for Design and Evaluation of a Neurally-controlled Powered Transfemoral Prosthesis
Published on: July 22, 2014
Stumble detection and classification for an intelligent transfemoral prosthesis
Brian E Lawson1, H Atakan Varol, Frank Sup
1Department of Mechanical Engineering, Vanderbilt University, Nashville, TN 37235, USA. brian.e.lawson@vanderbilt.edu
Summary
This study presents a real-time stumble detection system for intelligent lower limb prostheses using accelerometers. The system accurately identifies stumbles and classifies response types, enhancing prosthetic safety and function.
Area of Science:
- Biomedical Engineering
- Robotics
- Rehabilitation Technology
Background:
- Lower limb prostheses aim to restore mobility, but stumble detection remains a challenge.
- Intelligent prostheses require robust systems to identify and respond to unexpected gait deviations.
- Current methods may lack real-time processing capabilities for immediate stumble response.
Purpose of the Study:
- To develop and validate a real-time stumble detection approach for intelligent lower limb prostheses.
- To create an algorithm for classifying stumble responses into elevating or lowering types.
- To assess the accuracy of the proposed system in identifying and classifying stumbles.
Main Methods:
- Utilized accelerometers mounted on a transfemoral prosthesis to collect gait data.
- Collected stumble data from 10 healthy subjects under controlled experimental conditions.
- Developed and applied algorithms for real-time stumble detection and response classification.
Main Results:
- The proposed algorithms successfully detected all stumbles in real-time.
- The system accurately classified stumble responses as either elevating or lowering type.
- Achieved 100% accuracy in identifying 19 stumbles and 34 control strides.
Conclusions:
- The developed accelerometer-based system provides effective real-time stumble detection for lower limb prostheses.
- Accurate classification of stumble response types can inform prosthetic control strategies.
- This approach holds significant potential for improving the safety and adaptability of intelligent prosthetic devices.
