Autonomous Multi-agent Systems Using SVGS Camera Sensor for Lunar Surface Mobility Applications

Author: Alexander Frye, Department of Computer Science and Engineering

Mentor: Dr. Hadi Fekrmandi, Department of Mechanical Engineering

Design of a reliable mechanism of operation in autonomous multi-agent systems (MAS) demands several fundamental advances in distributed localization, guidance, navigation, and state estimation. Control and coordination of a team of mobile robots for cooperative functions in harsh and remote environments in space is an even greater challenge than the MAS systems on earth's surface. As opposed to dealing with single system dynamics, here the problem is the dynamics and consensus of the entire swarm state. Our goal is to study the Smartphone Video Guidance System (SVGS) camera system as a potential navigation camera sensor and create a framework to obtain relative and global states of multi-agent groups. The goal is for rovers to obtain a reliable estimation of the other agents through a decentralized state estimation method, targeted at robots searching for water-ice in the challenging domain of the lunar south pole.