Implementation of ensemble Kalman filter and consensus control for multi-robot motion system

Abstract

For decades, the development of robots has been very significant along with technological development. Robots used are to help complete human tasks. Initially, researchers conducted a lot of research related to single-robot motion systems. However, under certain conditions, the single-robot cannot help complete human tasks and cannot work in large-scale environments, so a multi-robot system is developed. A Multi-robot system is a robot in an environment unknown to work cooperatively for a particular purpose. To estimate motion in a multi-robot motion system, one of the data assimilation methods can be used, namely Ensemble Kalman Filter. The Ensemble Kalman Filter used to estimate various problems shaped model strongly nonlinear system and has shown that able to solve a dynamical system nonlinear model in the state of being big. After estimation, this paper also controls the movement of the robot. The control design used is consensus control based on graph theory and control theory. In consensus control, the connected graphs are intended towards a specific goal, namely towards the point of agreement. The results of the implementation of Ensemble Kalman Filter and consensus control on a multi-robot motion system obtained worked very well with very small error.