Using Octree Maps and RGBD Cameras to Perform Mapping and A* Navigation

Abstract

Ability to explore unknown environments andability to successfully navigate in these environments are two keyabilities of any truly autonomous mobile robot. In this article, anefficient implementation of mapping and navigation using onlyRGBD cameras is presented. The map created by Visual SLAMtechniques is converted into Octree structure, and successful A Navigation between two points in 3D is performed. Thecreated mapping system is able to operate in real-time even onsmall mobile robot and can be easily extended with new state-of the art techniques. Proposed way of octree encoding, due to itsminimum storage requirements, enables efficient distribution ofmap between multiple robots over the network. Presented pathplanningmethod with its optimizations runs in sufficient timeeven in higher resolution maps.