Leader-Follower Localization and Mapping using Range-Only Measurements

Abstract

A major part of the autonomous mobile robotics research is to address the localization and mapping problem of mobile robots using external sensory measurements. Extensive research has been conducted in the literature to solve this problem. However, a cost-effective and modular solution to the mapping and localization problem of a mobile robot operating in an indoor environment remains open. Addressing such a problem is even more challenging when the robot has to solve the leader-follower problem where neither leader nor follower robot states are known a priori. This work pertains to address this problem using range-only measurements from a set of networked wireless (radio) sensors placed in the robot's operating space (two-dimensional). A customized radio sensor is equipped with the follower robot. Since the locations/states of both robots (leader and follower) are unknown a priori, a conventional Extended Kalman Filter Simultaneous Localization and Mapping (EKF-SLAM) technique is employed for the follower robot to estimate its position and orientation, the position of the leader, and the positions of the radio sensors (i.e., building the map) placed in the robots' operating environment using range-only measurements from radio sensors. The proposed range-based approach for solving the leader-follower localization and mapping problem is validated using a commercially available robot simulator, V-REP (Virtual Robot Experimentation Platform).