Reinforcement Learning-Based Formation Pinning and Shape Transformation for Swarms

• DOI: 10.3390/drones7110673
• Publication Date: 2023-11-13T07:46:47Z
• Authors: Zhaoqi Dong; Qizhen Wu; Lei Chen
• Keywords: Flocking (texture); Adaptability; Drone; Collective Behavior; Cohesion (chemistry)

Swarm models hold significant importance as they provide the collective behavior of self-organized systems. Boids model is a fundamental framework for studying emergent in swarms It addresses problems related to simulating autonomous agents, such alignment, cohesion, and repulsion, imitate natural flocking movements. However, traditional often lack pinning adaptability quickly adapt dynamic environment. To address this limitation, we introduce reinforcement learning into solve problem disorder pinning. The aim approach enable drone effectively external environments. We propose method based on Q-learning network improve cohesion repulsion parameters achieve continuous obstacle avoidance maximize spatial coverage simulation scenario. Additionally, virtual leader coordination stability, reflecting leadership seen swarms. validate effectiveness method, demonstrate model’s capabilities through empirical experiments with swarms, show practicality RL-Boids framework.