Kinematic Modeling at the Ant Scale: Propagation of Model Parameter Uncertainties

Area of Science:

• Biomechanics
• Computational Biology
• Robotics

Background:

• Animal locomotion involves complex adaptations to environmental functions.
• Computational modeling aids understanding of locomotive patterns and body dynamics.
• Limitations of small-scale kinematic simulations remain unclear.

Purpose of the Study:

• Evaluate the impact of modeling uncertainties on small-scale kinematic simulations.
• Investigate sensitivity to joint geometry and marker position uncertainties.
• Enhance accuracy of multibody models for small, multi-legged animals.

Main Methods:

• Developed a multibody model of a *Messor barbarus* ant using X-ray micro-computed tomography scans.
• Estimated joint geometrical parameters from exoskeleton articular surfaces.
• Performed inverse kinematics simulations in OpenSim using high-speed video data and virtual markers.

Main Results:

• Assessed model sensitivity to joint geometrical parameters and marker position uncertainties via Monte Carlo simulations.
• The model demonstrated four times greater sensitivity to marker position perturbations than joint geometry.
• Identified key factors influencing the reliability of small-scale kinematic simulations.

Conclusions:

• Marker position uncertainty is a critical factor in small-scale kinematic simulations.
• Results inform locomotion studies of multi-legged animals, including insects and small vertebrates.
• Highlights the need for precise marker placement in biomechanical modeling of small organisms.