A Distributed Strategy for Cooperative Autonomous Robots Using Pedestrian Behavior for Multi-Target Search in the...

Area of Science:

• Robotics
• Artificial Intelligence
• Multi-agent Systems

Background:

• Multi-target search is a critical challenge for swarm robots.
• Existing methods often struggle with realistic constraints like limited communication and unknown environments.
• Efficient collision avoidance is essential for swarm coordination.

Purpose of the Study:

• To propose a novel swarm robotic pedestrian behavior (SRPB) model for multi-target search.
• To evaluate SRPB's performance under various realistic constraints.
• To enhance swarm robot efficiency in complex search scenarios.

Main Methods:

• Developed the swarm robotic pedestrian behavior (SRPB) model inspired by human pedestrian dynamics.
• Simulated multi-target search scenarios incorporating constraints: limited communication, time, unknown sources, arbitrary initial locations, and no central coordination.
• Evaluated performance metrics including time to find sources, number of located sources, and collision rates.

Main Results:

• SRPB demonstrated excellent stability and quick source-seeking capabilities.
• The model achieved a high number of located sources with a low collision rate.
• Performance was consistent across various experimental conditions, including different target signals, initial locations, population sizes, and target numbers.

Conclusions:

• The proposed SRPB model offers a robust and efficient solution for multi-target search in swarm robotics.
• SRPB effectively addresses realistic operational constraints, outperforming traditional approaches.
• This behavior-inspired model enhances swarm intelligence for complex environmental exploration and search tasks.