Moving-Distance-Minimized PSO for Mobile Robot Swarm

Area of Science:

• Robotics
• Artificial Intelligence
• Optimization Algorithms

Background:

• Particle Swarm Optimization (PSO) and mobile robot swarms are distinct swarm intelligence techniques.
• Their potential for synergistic application in real-world exploration and optimization is underexplored.
• Mobile robot swarms can perform PSO in physical spaces, but robot movement constraints (energy, time) are critical.

Purpose of the Study:

• To propose a novel Moving-Distance-Minimized PSO (MPSO) algorithm tailored for mobile robot swarms.
• To minimize the total travel distance of robots during the optimization process.
• To enhance the efficiency of robot swarms in exploring solution spaces by reducing energy and time costs.

Main Methods:

• Developed an MPSO algorithm that integrates mobile robot swarm movement with PSO principles.
• Utilized distances between current robot positions and next-generation particle positions to derive optimized paths.
• Focused on minimizing the cumulative distance traveled by all robots in the swarm.

Main Results:

• Experiments on 28 CEC2013 benchmark functions demonstrated MPSO's superiority over standard PSO.
• Achieved over 66% reduction in total moving distance for the robot swarm.
• Reduced the makespan (completion time) by nearly 70%.

Conclusions:

• The proposed MPSO effectively minimizes robot travel distance and energy consumption.
• MPSO offers significant improvements in efficiency and time reduction for robot swarm optimization tasks.
• The method maintains comparable optimization effectiveness to standard PSO while addressing practical robot movement constraints.