Communication and Control Algorithms for a Heterogenous Multi-agent System

Abstract

The SICOLET project is centered around a collaborative multi-agent system consisting of an ekranoplan, an unmanned aerial drone and a ground- based vehicle which is equipped with a mechanical manipulation device. All the agents are capable of a semi-autonomous operation, meaning, they can all be controlled manually via analog commands from the operator using a radio controller or by using a control algorithm, the operator giving the agents a set of instructions consisting of a set of waypoints and commands specific for every agent. Each agent is equipped with cameras, a wireless 5.8 GHz transmitter, a 2.4 GHz transceiver and a Raspberry Pi that allows localized image processing when the agent is unable to reach the command center via normal means of communication due to interferences from the environment. This paper focuses on the radio network and the algorithms that allow the agents to communicate with each other in the absence of a command signal from the command center. They are developed using Python and C++ programming languages.