Using a multi-robot system for improved path planning

Department of Electrical and Electronic Engineering, Stellenbosch University, South Africa

^* Corresponding author: cfisher@sun.ac.za

Abstract

Numerous terrestrial robotic platforms use computational power for path planning. These platforms typically use a vision system to identify obstacles and perform path planning. In cases where the vision systems are unable to function due to larger obstacles in the area, the paths are chosen as random functions of the given terrain to explore the environment and often results in missteps and moving away from the target location. In this research, we present a multi-robot system comprised of a terrestrial robot with a tethered aerial drone. By making use of the additive overhead view, the target location can be identified, and the exploration and path planning algorithms biased, subsequently reducing the computational cost, and creating a more efficient path planning approach. Focus has been placed on the control architecture of the system.