Energy efficient multi-robot exploration of an unknown 2-D environment
Abstract
This thesis investigates the problem of multi-robot exploration of an unknown environment. The goal of this study was to develop a method for multiple robots to successfully explore an unknown environment that minimized energy consumed due to travel. The method was based on an algorithmic approach that is characterized with decentralized communication, an iterative cost calculation, and utility based decision making. ^ For this work, energy formulas were developed using the kinematic equations, equations of motion, and kinetic energy equations. These energy formulas are used to calculate the energy required for translational and rotational motion. ^ The method was created by examining a method developed by Burgard [1], and modifying it. Using the energy calculations, the algorithm predicts the energy required to travel to exploration points. These explorations points are frontier goals which border unsensed areas. The algorithm determines which frontier goal and associated energy cost offers the robot greater utility and plans a path to travel to the goal. Once the goal is selected, the algorithm lowers the utility of the surrounding cells to entice robots to explore new territory. ^ The algorithm was tested using simulation software to test exploration effectiveness and cooperative characteristics. The energy calculations were tested using the iRobot Create and a camera control system. The findings of this thesis show that the algorithm is successful at exploring unknown environments, and that the energy formulas developed represent the energy required for motion well. ^
Subject Area
Engineering, Mechanical|Engineering, Robotics|Energy
Recommended Citation
Stephen Panasuik,
"Energy efficient multi-robot exploration of an unknown 2-D environment"
(2012).
Dissertations and Master's Theses (Campus Access).
Paper AAI1520177.
http://digitalcommons.uri.edu/dissertations/AAI1520177