A unified energy-optimality criterion predicts human navigation paths and speeds

Area of Science:

• Biomechanics
• Human Locomotion
• Robotics

Background:

• Human navigation involves complex movements, including turns, which are crucial for interacting with the physical environment.
• Existing models lack a unified theoretical framework to explain diverse non-straight-line human walking patterns.
• Understanding the energetic costs associated with turning is essential for predicting movement strategies.

Purpose of the Study:

• To develop a unified optimality criterion for predicting human walking paths and speeds in various scenarios.
• To investigate the metabolic cost of turning and its influence on trajectory selection.
• To demonstrate that energy optimization, rather than shortest path, governs human walking behavior.

Main Methods:

• Characterized the metabolic cost of turning as a function of turning radius and rate.
• Generalized the cost landscape to predict movement for complex trajectories, including holonomic walking (velocity differing from body orientation).
• Applied the optimality criterion to predict human movement in tasks like path following, turning in place, corridor navigation, and obstacle avoidance.

Main Results:

• Human walking speeds and paths align with predictions from the energy-optimality criterion.
• Individuals slow down when turning and avoid sharp turns, consistent with minimizing metabolic cost.
• The shortest path is often not the most energy-efficient, and humans select paths accordingly.

Conclusions:

• A unified optimality criterion successfully predicts diverse human walking behaviors, including non-straight paths and speeds.
• Energy efficiency is a primary factor influencing human gait and trajectory selection during navigation.
• The model provides a theoretical foundation for understanding human movement and can be extended to other populations and tasks.