Abstract:Computational walking simulations could facilitate the development of improved treatments for clinical conditions affecting walking ability. Since an effective treatment is likely to change a patient's foot-ground contact pattern and timing, such simulations should ideally utilize deformable foot-ground contact models tailored to the patient's foot anatomy and footwear. However, no study has reported a deformable modeling approach that can reproduce all six ground reaction quantities (expressed as three reaction force components, two center of pressure (CoP) coordinates, and a free reaction moment) for an individual subject during walking. This study proposes such an approach for use in predictive optimizations of walking. To minimize complexity, we modeled each foot as two rigid segments-a hindfoot (HF) segment and a forefoot (FF) segment-connected by a pin joint representing the toes flexion-extension axis. Ground reaction forces (GRFs) and moments acting on each segment were generated by a grid of linear springs with nonlinear damping and Coulomb friction spread across the bottom of each segment. The stiffness and damping of each spring and common friction parameter values for all springs were calibrated for both feet simultaneously via a novel three-stage optimization process that used motion capture and ground reaction data collected from a single walking trial. The sequential three-stage process involved matching (1) the vertical force component, (2) all three force components, and finally (3) all six ground reaction quantities. The calibrated model was tested using four additional walking trials excluded from calibration. With only small changes in input kinematics, the calibrated model reproduced all six ground reaction quantities closely (root mean square (RMS) errors less than 13 N for all three forces, 25 mm for anterior-posterior (AP) CoP, 8 mm for medial-lateral (ML) CoP, and 2 N·m for the free moment) for both feet in all walking trials. The largest errors in AP CoP occurred at the beginning and end of stance phase when the vertical ground reaction force (vGRF) was small. Subject-specific deformable foot-ground contact models created using this approach should enable changes in foot-ground contact pattern to be predicted accurately by gait optimization studies, which may lead to improvements in personalized rehabilitation medicine.

A multisegment computer simulation of normal human gait

Gait analysis of human locomotion based on motion capture system

Simulation of the Double Limb Support Phase of Human Gait

Simulation of Exoskeleton ZMP During Walking for Balance Control

Simulation of human gait with the aid of a simple mechanical model

Development of a Subject-Specific Foot-Ground Contact Model for Walking

Forward dynamic human gait simulation using a SLIP target model

3LP: A linear 3D-walking model including torso and swing dynamics

Design and validation of a foot-ankle dynamic simulator with a 6-degree-of-freedom parallel mechanism

The robustness of a second order sliding mode approach to human gait simulation

Human gait simulation with a neuromusculoskeletal model and evolutionary computation

Dynamic Simulation of Human Gait Model With Predictive Capability

A Multibody Dynamics Approach to Limit Cycle Walking.

A review of simulation methods for human movement dynamics with emphasis on gait

Human gait analysis based on OpenSim

A forward-inverse dynamics modeling framework for human musculoskeletal multibody system

An Interactive Simulator for Imposing Virtual Musculoskeletal Dynamics

Numerical Simulations and Experimental Human Gait Analysis Using Wearable Sensors

Biomechanical Analysis of Human Gait When Changing Velocity and Carried Loads: Simulation Study with OpenSim

Scalable Closed-Form Trajectories for Periodic and Non-Periodic Human-Like Walking

Simulation of human gait with body weight support: benchmarking models and unloading strategies