Virtual Velocity Obstacles
Abstract
This paper presents a novel approach to swarm navigation that combines hierarchical abstractions, flocking behaviors, and an efficient collision avoidance mechanism. Our main objective is to keep large groups of robots segregated while safely navigating in a shared environment. For this, we propose the Virtual Group Velocity Obstacle, which is an extension of the Velocity Obstacle concept for groups of robots. By augmenting velocity obstacles with flocking behaviors and hierarchical abstractions, we are able to navigate robotic swarms in a cohesive and smooth fashion. A series of simulations and real experimentswere performed and the results show the effectiveness of the proposed approach.