Human-swarn interaction with multi-robot coverage control in virtual reality environment

Abstract

This work presents an implementation of a human-swarm interaction applied in a coveragecontroltaskinnon-convexenvironments. Theoperatorsarecapableofinteracting with the robot swarm in a Virtual Reality (VR) environment, that allows a deep level of control, immersivity and intuitivity. On the VR application, the multi-robot interaction is abstracted in simple actions, allowing the user to interact the robot swarm by creating and removing obstacles, creating density distribution functions and altering the speed of robots. Simulations were implemented using ROS and Unity and demonstrated that robots could react to the changes in the environment introduced by the operators. Experiments were doneusingagroupofe-puckrobotsusingalocalizationsystemthatusesaugmentedreality tags. Additionally, a new algorithm for calculation of Voronoi diagrams was proposed. This algorithm, called Voro*, allows calculating local diagrams, enabling this algorithm to be applied in a large scale of robot groups with a better performance.