Particle filter in simultaneous localization and mapping (SLAM) using differential drive mobile robot

Simultaneous Localization and Mapping (SLAM) problem is a well-known problem in robotics, where a robot has to localize itself and map its environment simultaneously. Particle filter (PF) is one of the most adapted estimation algorithms for SLAM apart from Kalman filter (KF) and Extended Kalman Filter (EKF). In this work, particle filter algorithm has been successfully implemented using a simple differential drive mobile robot called e-puck. The performance of the algorithm implemented is analyzed via varied number of particles. From simulation, accuracy of the resulting maps differed according to the number of particles used. The Root Mean Squared Error (RMSE) of a larger number of particles is smaller compared to a lower number of particles after a period of time