EdgeConv-Based Spatial Difference Extraction of Plantar Pressure Improves Mediolateral Ground Reaction Force Estimation During Walking
Abstract
Falls and fractures are major concerns in super-aged societies, highlighting the need for gait assessment in daily living environments. Wearable footwear devices equipped with plantar pressure sensors and inertial measurement units (IMUs) enable remote gait monitoring without laboratory-based equipment; however, they cannot directly measure three-dimensional ground reaction forces (3D GRFs), and estimating the mediolateral component (Fx), which is related to balance control, remains challenging. This paper proposes a deep learning model that integrates Edge Convolution (EdgeConv), a graph neural network layer, to extract spatial features from plantar pressure data and improve Fx estimation. Gait experiments were conducted with 12 healthy participants walking on an instrumented treadmill, during which 3D GRFs and wearable sensor data from eight plantar pressure sensors and a six-axis IMU were synchronously recorded. The proposed model combines EdgeConv for spatial feature extraction with a Transformer for temporal feature learning and was evaluated using subject-independent six-fold group cross-validation. For Fx estimation, the proposed model achieved an NRMSE of 8.79 ± 1.26% and an R² of 0.76 ± 0.09, compared with 9.30 ± 1.52% and 0.73 ± 0.09 for the Transformer and 9.47 ± 1.77% and 0.71 ± 0.13 for the GCN. These results indicate that learning local spatial features with EdgeConv enhances mediolateral GRF estimation from wearable footwear sensor data.
Keywords: 3D Ground Reaction Forces (3D GRFs), EdgeConv, Walking Analysis, Plantar Pressure, IMU Sensor
DOI: 10.54941/ahfe1008308
Cite this paper
More from this volume
- AI-Enhanced Live-Streamed English Language Learning: Designing and Evaluating an Intelligent Platform for Chinese Learners
- "They Will Move": Confirmation Bias, Expectation Frames, and Premature Monitoring Drop Off in COLREG Encounters
- Automation transparency communication does not improve supervisory control performance, but shapes operator trust
- The influence of neuroticism on startle and surprise response: a flight simulator study
- EEG-Based Classification of Reward, Penalty, and Movement Events Across Four Video Games of Varying Complexity
- The Effects of Task Similarity and Time Pressure on Task Performance Following Interruptions in Complex Human-Computer Interaction Systems
- Trusting the Chain, Not the Black Box: A Provenance-Based Architecture for Human-Machine Teaming in Military Autonomy
- Design of a Simplified Humanoid Robot Face: Cultural Differences and Emotional Synchronization
- From Visual Observation to Auditory Interaction: A Human-Centered Tangible Music System Based on Physical Objects
- The Impact of Scarcity-Based Dark Patterns on User Performance, Perceived Darkness, and Workload in Subscription Services
- A Behavioral Observation Tool for Diagnosing Human–AI Interaction Profiles in Engineering Design Education
- The Mediated Moon: Semantic Mediation Between Gesture and Generative Image


AHFE Open Access