Strategies of locomotor collision avoidance

Area of Science:

• Human locomotion
• Robotics and biomechanics
• Human-computer interaction

Background:

• Collision avoidance is crucial for autonomous systems and human navigation.
• Previous research suggests a maximum smoothness strategy for trajectory planning in static environments.
• Dynamic environments introduce complexities not addressed by static models.

Purpose of the Study:

• To investigate collision avoidance strategies in dynamic, human-interactive scenarios.
• To determine if maximum smoothness applies to avoiding moving obstacles.
• To analyze locomotor path planning and execution in real-time, unpredictable situations.

Main Methods:

• Human participants navigated a space to avoid an orthogonally crossing pedestrian.
• Locomotor trajectories were recorded and analyzed for speed and path adjustments.
• Computational models simulated trajectories using a smoothness-maximization criterion.

Main Results:

• Participants prioritized maintaining a straight path and adjusted walking speed to avoid collision.
• Experimentally chosen trajectories were not globally optimal in smoothness.
• Locally smooth trajectories were maintained, but globally smoother paths deviating from a straight line were avoided.

Conclusions:

• The maximum smoothness strategy is not sufficient for collision avoidance in dynamic environments with moving obstacles.
• Locomotor path planning adapts to dynamic conditions, favoring path consistency over global smoothness.
• Future research should explore adaptive strategies for navigation in complex, interactive environments.