Simulating the effect of ankle plantarflexion and inversion-eversion exoskeleton torques on center of mass kinematics during walking

Nicholas A. Bianco, Steven H. Collins, C. Karen Liu, Scott L. Delp

TL;DR

This study uses detailed musculoskeletal simulations to map how specific ankle exoskeleton torques (plantarflexion, inversion, eversion) affect the body's center of mass during walking. The findings reveal that the type and timing of these torques significantly influence forward, backward, and sideways motion, and importantly, that intrinsic muscle properties can dampen these effects, informing the design of more effective balance-assistive exoskeletons.

What Problem Does This Paper Address?

Falls are a major cause of death and disability for older adults, often stemming from a loss of balance during walking. While exoskeletons hold promise for assisting individuals with neuromuscular deficits to improve balance, there's a significant knowledge gap regarding how applied exoskeleton torques at the ankle and subtalar joints translate into changes in the body's overall motion (center of mass kinematics). Understanding these relationships is crucial for designing exoskeletons that can effectively provide stabilizing torques without causing unintended or destabilizing movements.

Methods

The researchers employed a two-step simulation approach:

  1. Muscle-Driven Walking Simulations: They first created highly realistic simulations of walking using OpenSim Moco. These simulations were "tuned" to accurately replicate experimental motion capture and ground reaction force data from five healthy adults. This step established a baseline of normal walking.
  2. Exoskeleton Torque Simulation: In the second step, they introduced simulated exoskeleton torques (plantarflexion, inversion, eversion, and combinations) applied to the ankle and subtalar joints. These torques were active for 15% of the gait cycle, specifically during the stance phase (between foot-flat and toe-off). To understand the role of the body's own muscles, they ran two types of simulations:
  3. Muscle-Driven: The simulated exoskeleton torques acted on a model where muscle activations were still active and could adapt to the applied torques.
  4. Torque-Driven: The muscle activations were "frozen" to their state from the normal walking simulation, effectively removing the influence of intrinsic muscle properties on the response to exoskeleton torques.

By comparing these two simulation types, they could isolate the effect of the body's natural muscle responses. They then analyzed changes in the center of mass (CoM) velocity, acceleration, and position at specific points in the gait cycle.

Datasets

The study utilized experimental data from five healthy adults walking on a treadmill. This data included:

  • Marker trajectories (motion capture)
  • Ground reaction forces
  • Electromyography (EMG) signals from 11 muscles in the right leg

This data was used to create and validate the muscle-driven walking simulations. No new datasets were introduced.

Key Insights and Ideas

  • Torque Specificity: Different types of ankle torques have distinct effects. Plantarflexion torques primarily influence fore-aft and vertical CoM motion, while inversion/eversion torques are more dominant in controlling lateral CoM motion.
  • Timing Matters: The timing of when an exoskeleton applies torque during the gait cycle significantly alters its effect on CoM kinematics.
  • Muscle Damping Effect: The body's own muscles act as a natural "shock absorber," reducing the magnitude of CoM changes induced by exoskeleton torques. This means exoskeletons might need to apply larger torques than predicted by simulations that ignore muscle dynamics to achieve a desired effect.
  • Combined Torques: Applying combined torques (e.g., plantarflexion + eversion) results in kinematic changes that are a combination of the individual torque effects, but not always a simple sum.

Key Results

  • Plantarflexion Torques: Produced upward and backward CoM velocity changes in mid-stance, and upward and forward changes near toe-off.
  • Eversion/Inversion Torques: Primarily caused lateral/medial CoM velocity changes in mid-stance, respectively.
  • Muscle Influence: Torque-driven simulations (without active muscle response) showed larger CoM kinematic changes than muscle-driven simulations, sometimes by nearly a factor of two for transverse plane velocities.
  • Magnitude of Change: The maximum CoM velocity change observed was 6% of the walking speed.

Limitations and Future Work

  • Healthy Gait Only: The simulations focused on healthy, unperturbed walking. Future work should investigate these effects in individuals with neuromuscular deficits or during perturbed walking.
  • Short Simulation Window: Exoskeleton torques were simulated for only 15% of the gait cycle. Longer-duration simulations would require models that incorporate how the nervous system modulates muscle activity in response to exoskeletons.
  • Massless Actuators: The study did not account for the physical mass of the exoskeleton device, which could influence results.
  • Inter-Subject Variation: The study used a generic model scaled to subjects but did not explicitly model inter-subject variations in subtalar joint axis orientation.

Future work should incorporate human motor control models, explore longer simulation durations, include device mass, and extend the framework to other joints (hip, knee) and populations with gait pathologies. The authors also provide their simulation framework and results freely for further research.

Summary updated

Abstract

Abstract Walking balance is central to independent mobility, and falls due to loss of balance are a leading cause of death for people 65 years of age and older. Bipedal gait is typically unstable, but healthy humans use corrective torques to counteract perturbations and stabilize gait. Exoskeleton assistance could benefit people with neuromuscular deficits by providing stabilizing torques at lower-limb joints to replace lost muscle strength and sensorimotor control. However, it is unclear how applied exoskeleton torques translate to changes in walking kinematics. This study used musculoskeletal simulation to investigate how exoskeleton torques applied to the ankle and subtalar joints alter center of mass kinematics during walking. We first created muscle-driven walking simulations using OpenSim Moco by tracking experimental kinematics and ground reaction forces recorded from five healthy adults. We then used forward integration to simulate the effect of exoskeleton torques applied to the ankle and subtalar joints while keeping muscle excitations fixed based on our previous tracking simulation results. Exoskeleton torque lasted for 15% of the gait cycle and was applied between foot-flat and toe-off during the stance phase, and changes in center of mass kinematics were recorded when the torque application ended. We found that changes in center of mass kinematics were dependent on both the type and timing of exoskeleton torques. Plantarflexion torques produced upward and backward changes in velocity of the center of mass in mid-stance and upward and forward velocity changes near toe-off. Eversion and inversion torques primarily produced lateral and medial changes in velocity in mid-stance, respectively. Intrinsic muscle properties reduced kinematic changes from exoskeleton torques. Our results provide mappings between ankle plantarflexion and inversion-eversion torques and changes in center of mass kinematics which can inform designers building exoskeletons aimed at stabilizing balance during walking. Our simulations and software are freely available and allow researchers to explore the effects of applied torques on balance and gait.
BibTeX
@article{bianco2023simulating,
  author = {Nicholas A. Bianco and Steven H. Collins and C. Karen Liu and Scott L. Delp},
  title = {{Simulating the effect of ankle plantarflexion and inversion-eversion exoskeleton torques on center of mass kinematics during walking}},
  journal = {PLOS Computational Biology},
  year = {2023},
  url = {https://openreview.net/forum?id=QdH192pNVR},
}

Loading related content...

Figure