Conference Paper

Uncertain map making in natural environments

Lab. d'Autom. et d'Anal. des Syst., CNRS, Toulouse
DOI: 10.1109/ROBOT.1996.506847 Conference: Robotics and Automation, 1996. Proceedings., 1996 IEEE International Conference on, Volume: 2
Source: IEEE Xplore

ABSTRACT Building on previous work on incremental natural scene modelling
for mobile robot navigation, we focus in this paper on the problem of
representing and managing uncertainties. The environment is composed of
ground regions and objects. Objects (e.g., rocks) are represented by an
uncertain state vector (location) and a variance-covariance matrix.
Their shapes are approximated by ellipsoids. Landmarks are defined as
objects with specific properties (discrimination, accuracy) that permit
to use them for robot localization and for anchoring the environment
model. Model updating is based on an extended Kalman filter.
Experimental results are given that show the construction of a
consistent model over tens of meters

  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: This paper introduces a new terrain mapping method for mobile robots with a 2-D laser rangefinder. In the proposed method, an elevation map and a certainty map are built and used for the filtering of erroneous data. The filter, called Certainty Assisted Spatial (CAS) filter, first employs the physical constraints on motion continuity and spatial continuity to distinguish corrupted pixels (e.g., due to artifacts, random noise, or the "mixed pixels" effect) and missing data from uncorrupted pixels in an elevation map. It then removes the corrupted pixels and missing data, while missing data is filled in by a Weighted Median filter. Uncorrupted pixels are left intact so as to retain edges of objects. Our extensive indoor and outdoor mapping experiments demonstrate that the CAS filter has better performance in erroneous data reduction and map detail preservation than existing filters.
    Proc SPIE 01/2003; 5083:52-62.
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: This paper presents an obstacle avoidance system for the Segway Robotic Mobility Platform (RMP). The system consists of four main modules: terrain mapping, terrain traversability analysis, path planning, and motion control. The main sensor in our system is a forward/downward-looking 2-D Sick laser rangefinder. The terrain mapping module registers real- time laser range data into a grid-type elevation map. The traversal property of the elevation map is then analyzed by the traversability analysis module, which transforms the elevation map into a traversability map.The paper introduces a new con- cept called "traversability field histogram," which is used to transform the traversability map into a one-dimensional polar histogram. Finally, the path planning module determines the steering and velocity commands and sends them to the motion control module.
  • [Show abstract] [Hide abstract]
    ABSTRACT: This paper addresses the problem of globally- consistent localization and mapping simultaneously for au- tonomous mobile robots in an unknown and unstructured D environment by sensor fusion. It belongs to the research area of simultaneous localization and mapping (SLAM) in a mobile robot community. SLAM problem could be solved with filter- based method, and graph-based method. The basic requirement for these methods is that the measurement for the landmarks at every step should be reliable and will provide enough useful information. In real application, it is impossible to ensure the sen- sor, such as camera, to obtain enough landmark measurements at every step. In this paper, we present new simple measurement architecture and an algorithm for a globally-consistent SLAM solution with senor fusion, which could be used by a mobile robot in some environment where the landmark is very spare. Outdoor environment experiment has shown that this method is reliable and possible. Index Terms—Sensor fusion, SLAM, Robot localization, Stereo camera.