Swarm formation morphing for congestion-aware collision avoidance.

Area of Science:

• Robotics
• Multi-agent systems
• Swarm intelligence

Background:

• Swarm formations face challenges in obstacle avoidance.
• Overpopulation on one side of an obstacle leads to delays and increased energy consumption.
• Efficient re-formation after obstacle evasion is crucial for mission success.

Purpose of the Study:

• To present a novel methodology for optimal swarm formation distribution during obstacle evasion.
• To reduce agent waiting delays and minimize overall mission time and energy consumption.
• To develop distinct strategies for the disturbance (evasion) and convergence (re-formation) phases.

Main Methods:

• Dividing the problem into a disturbance phase (optimal morphing to avoid obstacles) and a convergence phase (resuming formation shape).
• Developing a methodology for the disturbance phase that tests formation morphing combinations using trajectory, velocity, and coordinate data.
• Utilizing a thin-plate splines (TPS) inspired temperature function minimization for the convergence phase to optimally restore the formation.

Main Results:

• The proposed methodology effectively manages swarm distribution around obstacles.
• The approach significantly reduces agent waiting delays and optimizes traversal time.
• Experimental results demonstrate substantial energy savings compared to traditional methods in simulated scenarios.

Conclusions:

• The novel methodology provides an optimal solution for swarm formation obstacle avoidance.
• The two-phase approach (disturbance and convergence) ensures efficient evasion and timely re-formation.
• The method leads to reduced energy consumption and improved overall mission efficiency for robotic swarms.