Improving Multi-Agent SLAM with Information Theory and a Gimbal

The localization and mapping of a robot and its immediate environment is a fundamental problem for any robot navigating in an unknown environment, this problem is commonly known as Simultaneous Localization and Mapping or SLAM [3, 4]. Algorithms that perform SLAM often use sensors on-board the robot to estimate the robot state which commonly

includes the position and orientation, but may include other quantities such as velocity, sensor biases and calibration parameters, all while producing a map of the environment to which the robot is operating in. There is extensive literature and advancements in SLAM in recent decades, with the popularity of SLAM connected to the fact that manually built

map of the environment can be omitted, and showing robot operation is possible without adding artificial beacons for robot localization. Research area: Smart Robotics