Contact

Need help?

For technical questions about Moticon products

Contact Support

Want to say hello?

Drop us a message for general questions or inquiries

Write Us

Up for a live demo?

See our products in action and ask your questions

Book a Demo

Interested in prices?

Get the official price list or an individual quote

Get a Quote

Always just a call away

+49 89 2000 301 60

Signal Processing and Machine Learning for Sensor Systems

Impact of PCA Pre-Normalization Methods on Ground Reaction Force Estimation Accuracy

Amal Kammoun, Philippe Ravier, Olivier Buttelli

PRISME Laboratory, University of Orleans, Orleans, France

Download PDF >

DOI Link >

Keywords

insole measurement, force plate measurement, GRF component estimation, normalization methods, machine learning, PCA pre-normalization

Abstract

Ground reaction force (GRF) components can be estimated using insole pressure sensors. Principal component analysis in conjunction with machine learning (PCA-ML) methods are widely used for this task. PCA reduces dimensionality and requires pre-normalization. In this paper, we evaluated the impact of twelve pre-normalization methods using three PCA-ML methods on the accuracy of GRF component estimation. Accuracy was assessed using laboratory data from gold-standard force plate measurements. Data were collected from nine subjects during slow- and normal-speed walking activities. We tested the ANN (artificial neural network) and LS (least square) methods while also exploring support vector regression (SVR), a method not previously examined in the literature, to the best of our knowledge. In the context of our work, our results suggest that the same normalization method can produce the worst or the best accuracy results, depending on the ML method. For example, the body weight normalization method yields good results for PCA-ANN but the worst performance for PCA-SVR. For PCA-ANN and PCA-LS, the vector standardization normalization method is recommended. For PCA-SVR, the mean method is recommended. The final message is not to define a normalization method a priori independently of the ML method.

Moticon's Summary

The paper discusses the importance of normalization in machine learning (ML) and principal component analysis (PCA) for estimating ground reaction force (GRF) components using sensor insole data. The study evaluates twelve normalization methods and three ML techniques (artificial neural networks (ANN), least squares (LS), and support vector regression (SVR)) to determine the most effective combinations. The experimental setup involved nine healthy male subjects performing walking tasks, with data collected using Moticon sensor insoles and force plates. Results indicate that the PCA-ANN method with variable standardization (VS) normalization generally provided the best estimation accuracy for GRF components, followed by PCA-SVR and PCA-LS methods. The study highlights the variability in effectiveness of normalization methods depending on the ML technique used, challenging previous assumptions that certain normalization methods, like Z-score, are universally optimal.

Book a free online demo or use the contact form to get in touch

Newsletter

Subscribe to our newsletter for regular updates

Overview

Tests

Sensor Insoles

Accessories

with OpenGo

Overview

App

Services

Sensor Insoles

Software

Accessories

with ReGo

Need help?

For technical questions about Moticon products

Want to say hello?

Drop us a message for general questions or inquiries

Up for a live demo?

See our products in action and ask your questions

Interested in prices?

Get the official price list or an individual quote

Always just a call away

Signal Processing and Machine Learning for Sensor Systems

Impact of PCA Pre-Normalization Methods on Ground Reaction Force Estimation Accuracy

Amal Kammoun, Philippe Ravier, Olivier Buttelli

PRISME Laboratory, University of Orleans, Orleans, France

Keywords

insole measurement, force plate measurement, GRF component estimation, normalization methods, machine learning, PCA pre-normalization

Abstract

Moticon's Summary

Overview

Sensor Insoles

Tests

Accessories

with OpenGo

Overview

Sensor Insoles

App

Software

Services

Accessories

with ReGo

Select your desired system

Have a general inquiry?

The form was sent successfully.

Stay one step ahead!

Select your desired system

Get support

Check our FAQ database for answers to frequently asked questions