Analysis of Different Feature Selection Criteria Based on a Covariance Convergence Perspective for a SLAM Algorithm

Fernando A Auat Cheein,Ricardo Carelli

doi:10.3390/s110100062

Abstract

This paper introduces several non-arbitrary feature selection techniques for a Simultaneous Localization and Mapping (SLAM) algorithm. The feature selection criteria are based on the determination of the most significant features from a SLAM convergence perspective. The SLAM algorithm implemented in this work is a sequential EKF (Extended Kalman filter) SLAM. The feature selection criteria are applied on the correction stage of the SLAM algorithm, restricting it to correct the SLAM algorithm with the most significant features. This restriction also causes a decrement in the processing time of the SLAM. Several experiments with a mobile robot are shown in this work. The experiments concern the map reconstruction and a comparison between the different proposed techniques performance. The experiments were carried out at an outdoor environment composed by trees, although the results shown herein are not restricted to a special type of features.

Highlights

This paper addresses the problem of feature selection within a feature-based simultaneous localization and mapping (SLAM) algorithm
The features of the environment are represented by blue triangles; the path traveled by the robot is a solid black line and the path estimated by the SLAM is a solid magenta line
This paper has presented several non-heuristic feature selection criteria for an Extended Kalman filter (EKF)-SLAM algorithm

Summary

Introduction

This paper addresses the problem of feature selection within a feature-based simultaneous localization and mapping (SLAM) algorithm. The most widely used by the scientific community is the Extended Kalman filter (EKF) [1,3,4,5,6] solution and its derived filters, such as the Unscented Kalman filter (UKF) [3] and the Extended Information filter (EIF) [7,8] In these filters, the SLAM system state, composed by the robot’s pose and the map of the environment, it is modeled as a Gaussian random variable. The feature extraction process determines the model associated with the environment and the map derived from the SLAM system state. This paper introduces several non-heuristic criteria to select the most significant features from the environment to be used in the correction stage of the SLAM algorithm. The feature selection criteria presented are not restricted to the type of features used within the SLAM, an EKF-SLAM with point-based features is used to show the performance of each proposal

Related Work

9: Let Nt be set of the observed features

Experimental Results

SLAM Results

Conclusions