Using the CONDENSATION algorithm for robust, vision-based mobile robot localization

  title={Using the CONDENSATION algorithm for robust, vision-based mobile robot localization},
  author={Frank Dellaert and Wolfram Burgard and Dieter Fox and Sebastian Thrun},
  journal={Proceedings. 1999 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (Cat. No PR00149)},
  pages={588-594 Vol. 2}
  • F. Dellaert, W. Burgard, S. Thrun
  • Published 23 June 1999
  • Computer Science
  • Proceedings. 1999 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (Cat. No PR00149)
To navigate reliably in indoor environments, a mobile robot must know where it is. This includes both the ability of globally localizing the robot from scratch, as well as tracking the robot's position once its location is known. Vision has long been advertised as providing a solution to these problems, but we still lack efficient solutions in unmodified environments. Many existing approaches require modification of the environment to function properly, and those that work within unmodified… 

Figures from this paper

Vision-based global localization and mapping for mobile robots
Experiments show that global localization can be achieved accurately using the scale-invariant landmarks, and the approach of pairwise submap alignment with backward correction in a consistent manner produces a better global 3-D map.
Active Appearance-Based Robot Localization Using Stereo Vision
An entropy-based criterion for action selection that can be efficiently evaluated in the probabilistic localization system is introduced and a novel Expectation-Maximization algorithm to determine the principal components of a data set including missing values is described and applied to depth maps.
Approaches to Mobile Robot Localization in Indoor Environments
An extensively tested low-complexity, robust and accurate pose tracking method is presented which utilizes the minimalistic model in combination with a laser sensor, based on the ideas of Multiple Hypothesis Tracking.
Markov Localization for Mobile Robots in Dynamic Environments
A version of Markov localization which provides accurate position estimates and which is tailored towards dynamic environments, and includes a filtering technique which allows a mobile robot to reliably estimate its position even in densely populated environments in which crowds of people block the robot's sensors for extended periods of time.
Omnidirectional vision scan matching for robot localization in dynamic environments
An omnidirectional camera mounted on a mobile robot is used to perform a sort of scan matching, which finds the distances of the closest color transitions in the environment, mimicking the way laser rangefinders detect the closest obstacles.
Mobile robot localization using a Kalman filter and relative bearing measurements to known landmarks
Although this research requires that the landmark locations be known, in contrast to research in simultaneous localization and mapping, the results are still useful in an industrial setting where placing known landmarks would be acceptable.
Probabilistic cooperative localization and mapping in practice
A probabilistic framework for the reduction in the uncertainty of a moving robot pose during exploration by using a second robot to assist and a Monte Carlo Simulation technique is employed in order to model and reduce the accumulated odometric error.
A Visual-Sensor Model for Mobile Robot Localisation
A probabilistic sensor model for camera-pose estimation in hallways and other known structured environments that is based on feature matching techniques that are simpler than state-of-the-art photogrammetric approaches and sufficiently fast to allow for sensor fusion.
Omnivision-based KLD-Monte Carlo Localization
Mobile Robot Localization and Mapping with Uncertainty using Scale-Invariant Visual Landmarks
A vision-based mobile robot localization and mapping algorithm, which uses scale-invariant image features as natural landmarks in unmodified environments to localize itself accurately and build a map of the environment.


Integrating global position estimation and position tracking for mobile robots: the dynamic Markov localization approach
  • W. BurgardAndrcas DerrD. FoxA. Cremers
  • Computer Science
    Proceedings. 1998 IEEE/RSJ International Conference on Intelligent Robots and Systems. Innovations in Theory, Practice and Applications (Cat. No.98CH36190)
  • 1998
The dynamic Markov localization technique is presented as a uniform approach to position estimation, able to globally estimate the position of the robot, to efficiently track its position whenever the robot's certainty is high, and to detect and recover from localization failures.
Position Estimation for Mobile Robots in Dynamic Environments
Extensions to Markov localization algorithms enabling them to localize mobile robots even in densely populated environments are proposed and implemented, demonstrating that this approach is able to accurately estimate the robot's position in more than 98% of the cases even in such highly dynamic environments.
Active Markov localization for mobile robots
AMOS: comparison of scan matching approaches for self-localization in indoor environments
  • J.-S. GutmannC. Schlegel
  • Computer Science
    Proceedings of the First Euromicro Workshop on Advanced Mobile Robots (EUROBOT '96)
  • 1996
This paper describes results from evaluating different self-localization approaches in indoor environments for mobile robots based on 2D laser scans and an odometry position estimate and shows that the position error can be kept small enough to perform navigation tasks.
Image-based prediction of landmark features for mobile robot navigation
Methods for direct image-based prediction of point and line features for a mobile system operating on a planar surface are discussed for efficient and with sufficient accuracy to ensure robust acquisition of navigational landmarks.
Topological Mapping for Mobile Robots Using a Combination of Sonar and Vision Sensing
This paper uses the robot's sonar sensors to determine where to capture images and use cues extracted from those images to help perform place recognition, and combines a sonar-based definition of distinctive places with visual information using a simple Bayesian network.
A Distributed Model for Mobile Robot Environment-Learning and Navigation
A distributed method for mobile robot navigation, spatial learning, and path planning is presented and the main issues addressed are: distributed, procedural, and qualitative representation and computation, emergent behaviors, dynamic landmarks, minimized communication.
Globally Consistent Range Scan Alignment for Environment Mapping
The problem of consistent registration of multiple frames of measurements (range scans), together with therelated issues of representation and manipulation of spatialuncertainties are studied, to maintain all the local frames of data as well as the relative spatial relationships between localframes.
Probabilistic Robot Navigation in Partially Observable Environments
First results are reported on first results of a research program that uses par tially observable Markov models to robustly track a robots location in office environments and to direct its goal-oriented actions.
Navigating Mobile Robots: Systems and Techniques
This is a survey of the state-of-the-art in sensors, systems, methods and technologies utilized by a mobile robot to determine its position in the environment. The many potential "solutions" are