Hybrid Motion Planner Integrating Global Voronoi Diagrams and Local Velocity Obstacle Method

Abstract

Global and local path planning are a classic research field in robotics and are key elements for the development of automated warehouse. Velocity Obstacle (VO) and Voronoi Diagrams (VD) are well known methods in autonomous navigation. VO provides for each autonomous vehicle a set of available velocities ensuring collision-free trajectory for multi-robot systems, whereas VD is a geometrical space subdivision used to compute paths furthest away from obstacles. In this paper we will focus on the design of a distributed navigation architecture aiming at integrating to integrate VO, as a local planner and VD as a global planner, to provides efficient collision-free and safe maneuver for each agent belonging to the fleet. The approach is validated in simulation using ROS and Gazebo as a virtual robot development environment.