A Multi-MAV System for the Autonomous Elimination of Multiple Targets in the MBZIRC 2020 Competition

Abstract

The Mohamed Bin Zayed International Robotics Challenge (MBZIRC) is a prestigious, biennial competition aimed at furthering the state of the art in the field of autonomous robotics. In this paper, we present our solution to one of the tasks in the MBZIRC 2020 competition, which design won second place in Challenge 1 and first place in the Grand Challenge of the competition. This paper focuses specifically on the popping task of multiple balloons by multiple Micro Aerial Vehicles (MAVs). This task required a rapid and robust performance to compete with systems from other expert robotic teams from around the world. In this task, a team of autonomous MAV’s had to seek and attack several balloons positioned throughout the competition arena. The novel fast autonomous searching for multiple targets in 3D, their reliable detection, precise relative state estimation, and agile motion planning algorithms are presented in this paper, together with an application for general tasks of 3D target capturing. With a primary focus on reliability, the methods reported in this paper and the entire, complex multi-agent system were successfully verified in both extreme conditions of the desert and the MBZIRC competition. An evaluation of the proposed methods using data from the competition and additional separate datasets is presented. The relevant code of our implementation has been made publicly available for the robotics community.