This study aims to contribute a comparison of various simultaneous localization and mapping (SLAM) algorithms that have been proposed in literature. The performance of Extended Kalman Filter (EKF) SLAM, Unscented Kalman Filter (UKF) SLAM, EKF-based FastSLAM version 2.0, and UKF-based FastSLAM (uFastSLAM) algorithms are compared in terms of accuracy of state estimations for localization of a robot and mapping of its environment. The algorithms were run using the same type of robot on Player/Stage environment. The results show that the UKF-based FastSLAM has the best performance in terms of accuracy of localization and mapping. Unlike most of the previous applications of FastSLAM in literature, no waypoints are used in this study.

A comparison of EKF, UKF, FastSLAM2.0, and UKF-based FastSLAM algorithms

A global robust iterative learning position control for current-fed permanent magnet step motors

Brief paper: An adaptive learning regulator for uncertain minimum phase systems with undermodeled unknown exosystems

FastSLAM is a framework which solves the problem of simultaneous localization and mapping using a Rao-Blackwellized particle filter. Conventional FastSLAM is known to degenerate over time in terms of accuracy due to the particle depletion in resampling phase. In this work, a new FastSLAM framework is proposed to prevent the degeneracy by particle cooperation. First, after resampling phase, a target that represents an estimated robot position is computed using the positions of particles. Then, particle swarm optimization is performed to update the robot position by means of particle cooperation. Computer simulations revealed that the proposed framework shows lower RMS error in both robot and feature positions than conventional FastSLAM.

/pdf/pso-fastslam-an-improved-fastslam-framework-using-particle-5d6fwiq5s6.pdf

PSO-FastSLAM: An improved FastSLAM framework using particle swarm optimization

Learning control for nonlinear systems in output feedback form

A FastSLAM approach to the SLAM problem is considered in this paper. An improvement to the classical FastSLAM algorithm has been obtained by replacing the Extended Kalman Filters used in the prediction step and in the feature update with Unscented Kalman Filters and by introducing an adaptive selective resampling. The simulations confirm the effectiveness of the proposed modifications.

/pdf/a-fastslam-algorithm-based-on-the-unscented-filtering-with-2lpnt602gn.pdf

A FastSLAM algorithm based on the unscented filtering with adaptive selective resampling

The output tracking problem of periodic reference signals (of known period) for single-input single-output observable minimum phase uncertain linear time-invariant systems with unitary relative degree is considered. A continuous global iterative learning control via output error feedback is designed which guarantees closed loop boundedness and asymptotic output tracking, thus improving the L 2 convergence achieved in [6]. Its closed loop performances are then compared with those obtained by using the adaptive learning control in [2] and the adaptive regulator in [5]: in particular the effects of finite memory implementation, measurement noise and uncertainties in the period of the output reference signal are explicitly taken into account in the simulations.

Manuel Cugliari

Papers

A FastSLAM algorithm based on the unscented filtering with adaptive selective resampling

Global learning controls for uncertain relative degree one linear systems: a comparative study