Conference Paper

Rao-Blackwellized Particle Filtering for Mapping Dynamic Environments

Sibley Sch. of Mech. & Aerosp. Eng., Cornell Univ., Ithaca, NY
DOI: 10.1109/ROBOT.2007.364071 Conference: Robotics and Automation, 2007 IEEE International Conference on
Source: IEEE Xplore

ABSTRACT A general method for mapping dynamic environments using a Rao-Blackwellized particle filter is presented. The algorithm rigorously addresses both data association and target tracking in a single unified estimator. The algorithm relies on a Bayesian factorization to separate the posterior into: 1) a data association problem solved via particle filter; and 2) a tracking problem with known data associations solved by Kalman filters developed specifically for the ground robot environment. The algorithm is demonstrated in simulation and validated in the real world with laser range data, showing its practical applicability in simultaneously resolving data association ambiguities and tracking moving objects.

0 Followers
 · 
71 Views
  • [Show abstract] [Hide abstract]
    ABSTRACT: The article proposes a solution to map-based self-localization for an autonomous robot operating in cluttered and crowded environments. To detect features for localization, 2D laser range-finders traditionally scan a plane parallel to the floor. This work hypothesizes the existence of a “low frequency cross-section” of the 3D Workspace where cluttered and dynamic environments become “more regular” and “less dynamic”. The contribution of the article is twofold. First, an “unevenness index” U is introduced to quantitatively measure the complexity of the environment as it would be perceived if the laser range-finder were located at different heights from the floor. The article shows that, by choosing the laser scanning plane to statistically minimize U (in most cases, above the heads of people), it is possible to deal more efficiently with non-linearities in the measurement model, moving objects and occluded features. Second, it is demonstrated that, when adopting an extended Kalman filter for position tracking (a very common and widely used technique in real-world scenarios), the a posteriori covariance of the estimated robot pose converges faster, on average, when U is lower, which leads to better localization performance. Experimental results show hours of continuous robot operation in real-world, cluttered and crowded environments.
    Journal of Intelligent and Robotic Systems 11/2012; 68(2). DOI:10.1007/s10846-012-9673-x · 0.81 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Many problems of state estimation in structural dynamics permit a partitioning of system states into nonlinear and conditionally linear substructures. This enables a part of the problem to be solved exactly, using the Kalman filter, and the remainder using Monte Carlo simulations. The present study develops an algorithm that combines sequential importance sampling based particle filtering with Kalman filtering to a fairly general form of process equations and demonstrates the application of a substructuring scheme to problems of hidden state estimation in structures with local nonlinearities, response sensitivity model updating in nonlinear systems, and characterization of residual displacements in instrumented inelastic structures. The paper also theoretically demonstrates that the sampling variance associated with the substructuring scheme used does not exceed the sampling variance corresponding to the Monte Carlo filtering without substructuring.
    Probabilistic Engineering Mechanics 10/2012; 30:89-103. DOI:10.1016/j.probengmech.2012.06.005 · 1.46 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Simultaneous localization and consistent mapping in dynamic environments is a fundamental and unsolved problem in the mobile robotics community. Most of the algorithms for this problem heavily rely on discriminating dynamic objects from static objects. Because these recursive filters based discrimination algorithms always have lag before the model selection parameters converge to the steady states, they have a period of time that the filter could identify a dynamic target as static or vice versa. Mis-classifications decrease precision and consistence, and induce filter divergence.A brain interacts with dynamic environments. The biological basis of this adaptability is provided by the connectivity and the dynamic properties of neurons. Biologically inspired by the adaptability, the paper proposes a shunting STM (Short Term Memory) based method to solve the simultaneous localization and consistent mapping problem, especially in dynamic environments. The proposed method utilizes a shunting STM neural network to represent environments and to probabilistically reflect the probability of existence of an object; it adapts a scan matching scheme to localize robot based on the map representation. Dynamic properties of the neural network are used to reflect environmental changes, therefore, the proposed method does not require explicit discrimination of objects. As a result, the proposed method does not have the lag of convergence, and it has high utilization ratio of observation information. Theoretical analyses in the paper show the proposed method has Lyapunov stability and its computational complexity does not depend on the size of the environment. The paper compares the proposed method with the classification based Extend Kalman Filter on a classical outdoor dataset, in simulated environments and in real indoor environments. The results show the proposed method outperforms the classification based EKF on precision and consistence in both static environments and dynamic environments.
    Neurocomputing 03/2013; 104:170–179. DOI:10.1016/j.neucom.2012.10.011 · 2.01 Impact Factor