SLAM based on Adaptive SVSF for Cooperative Unmanned Vehicles in Dynamic environment

Abstract

Abstract Autonomous navigation in unknown environments has attracted a special attention from the robotic community over the last three decades. In order to gather accurate information about the environment, wheeled mobile robots are equipped with a variety of sensors (e.g. laser, vision, sonar, GPS, etc.), allowing accurate localization and reconstruction of reliable and consistent representations of the environment. When neither the location of the robot nor the map of its surrounding known, localization and mapping become highly interdependent tasks that must be performed concurrently. This problem is known as Simultaneous Localization and Mapping (SLAM). Many techniques have been developed to make the utmost of the robot sensors to solve the SLAM problem. A solution to SLAM in dynamic environments would expand robotic applications and open up a vast range of potential applications. In this context, we propose the adaptive smooth variable structure filter (ASVSF) based approach to solve the SLAM problem and reconstruct a reliable 2D environment representation. The proposed algorithm is validated in real-world and the obtained results confirm the efficiency and robustness of our approach in dynamic environment.