Novel Mobile Robot Simultaneous Loclization and Mapping Using Rao-Blackwellised Particle Filter

Abstract

This paper presents the novel method of mobile robot simultaneous localization and mapping (SLAM), which is implemented by using the Rao-Blackwellised particle filter (RBPF) for monocular vision-based autonomous robot in unknown indoor environment. The particle filter is combined with unscented Kalman filter (UKF) to extending the path posterior by sampling new poses that integrate the current observation. The landmark position estimation and update is implemented through the unscented transform (UT). Furthermore, the number of resampling steps is determined adaptively, which seriously reduces the particle depletion problem. Monocular CCD camera mounted on the robot tracks the 3D natural point landmarks, which are structured with matching image feature pairs extracted through Scale Invariant Feature Transform (SIFT). The matching for multi-dimension SIFT features which are highly distinctive due to a special descriptor is implemented with a KD-Tree in the time cost of O(log2 N ). Experiments on the robot Pioneer3 in our real indoor environment show that our method is of high precision and stability.

A polynomial based SLAM algorithm for mobile robots using ultrasonic sensors - Experimental results

“…The authors solve the problem starting from an unknown robot initial position and using information from laser sensors through an Extended Kalman filter (EKF). In [4] the authors present a method to perform SLAM using the Rao-Blackwellised particle filter (RBPF) for monocular vision-based autonomous robot in an unknown indoor environment. In this work the particle filter is combined with an Unscented Kalman filter (UKF) and the environment mapping is implemented through the unscented transform.…”

“…In [7] the authors use sonar measurements along with a particle filter to solve the SLAM problem in non-static environments. In most of these works ( [2], [4], [3]) the authors use a scanner rangefinder sensor to solve the SLAM problem and thus the proposed solutions rely on accurate and dense measurements. Moreover, very often the SLAM algorithms assume to have at least some information about the environment (e.g.…”

Robot vision for the visually impaired

“…This approach has been proven to be quite successful in navigation for autonomous robotic systems [24,21] and urban/landscape modeling [26]. Vision based SLAM systems have also been proposed for wearable applications such as augmented reality [17].…”

Link Graph and Feature Chain Based Robust Online SLAM for Fully Autonomous Mobile Robot Navigation System Using Sonar Sensors

“…For correcting the angle assumption of the robot we take a weighted sum of angle difference from the roots (not with the associated links), and find the robot's orientation, which will minimize equation (4). Where arg_min returns the value of argument which will minimize the quantity inside the bracket, n is the number of associated features' instances.…”

“…But these approaches are basically dependent on the assumption that the inputs from the sensors are reliable and consistent. Generally a combination of sensors like vision and sonar [1] are used or a laser sensor [2,3] is used or camera [4] is used. But very few authors [5,6] have addressed the SLAM problem in indoor environment with only sonar sensors.…”