Implementation of Rao-Blackwellized Particle Filter on LIMO ROS2

Simultaneous Localization and Mapping (SLAM) is crucial for autonomous mobile robots navigating unknown environments. Traditional methods like Extended Kalman Filters and Particle Filters manage uncertainties but often falter in dynamic settings. This article discusses the implementation of the Rao-Blackwellized Particle Filter (RBPF) in the SLAM system of the LIMO ROS2 platform. By integrating RBPF into the default ROS2 framework, we decompose the SLAM problem into simpler sub-problems, enhancing computational efficiency and estimation accuracy. Results show that the implemented system effectively handles dynamic environments and noisy data, as demonstrated through simulations and real-world experiments with the LIMO ROS2 robot.