A particle filter for monocular vision-aided odometry

Abstract

Abstract-We propose a particle filter-based algorithm for monocular vision-aided odometry for mobile robot localization. The algorithm fuses information from odometry with observations of naturally occurring static point features in the environment. A key contribution of this work is a novel approach for computing the particle weights, which does not require including the feature positions in the state vector. As a result, the computational and sample complexities of the algorithm remain low even in feature-dense environments. We validate the effectiveness of the approach extensively with both simulations as well as real-world data, and compare its performance against that of the extended Kalman filter (EKF) and FastSLAM. Results from the simulation tests show that the particle filter approach is better than these competing approaches in terms of the RMS error. Moreover, the experiments demonstrate that the approach is capable of achieving good localization accuracy in complex environments.

Consistency Analysis and Improvement of Vision-aided Inertial Navigation

“…Every time a new IMU measurement is received, the state estimate is propagated using numerical integration of (10)- (15). In order to derive the covariance propagation equation, we compute the discrete-time state transition matrix, Φ k +1,k , from time-step t k to t k +1 , as the solution to the following matrix differential…”

“…Numerous VINS approaches have been presented in the literature, including methods based on the extended Kalman filter (EKF) [8]- [11], the unscented Kalman filter [12], and batch least squares (BLS) [13]. Nonparametric estimators, such as the particle filter, have also been applied to visual odometry (e.g., [14], [15]). However, these have focused on the simplified problem of estimating the pose of a vehicle whose motion is constrained to 2-D, since the number of particles required is exponential in the size of the state vector.…”

Camera-IMU-based localization: Observability analysis and consistency improvement

“…Non-parametric estimators, such as the Particle Filter (PF), have also been used for visual-inertial odometry (e.g. Durrie et al (2009); Yap et al (2011)). However, these have focused on the reduced problem of estimating a 2D robot pose, since the number of particles required is exponential in the size of the state vector.…”

Towards Consistent Vision-Aided Inertial Navigation

“…Non-parametric estimators, such as the Particle Filter (PF), have also been applied to visual odometry (e.g., [6,33]). However, these have focused on the simplified problem of estimating the 2D robot pose since the number of particles required is exponential in the size of the state vector.…”