Perceptual organization and the representation of natural form
Artificial Intelligence
Object recognition by computer: the role of geometric constraints
Object recognition by computer: the role of geometric constraints
A Method for Registration of 3-D Shapes
IEEE Transactions on Pattern Analysis and Machine Intelligence - Special issue on interpretation of 3-D scenes—part II
Perception as Bayesian inference
Perception as Bayesian inference
The spatial semantic hierarchy
Artificial Intelligence
Globally Consistent Range Scan Alignment for Environment Mapping
Autonomous Robots
Spatial Cognition and Computation
Fast Line and Rectangle Detection by Clustering and Grouping
CAIP '97 Proceedings of the 7th International Conference on Computer Analysis of Images and Patterns
Spatial Cognition IVReasoning, Action, Interaction: International Spatial Cognition 2004, Frauenchiemsee, Germany, October 11-13, 2004, Revised Selected ... / Lecture Notes in Artificial Intelligence)
Stanley: The robot that won the DARPA Grand Challenge: Research Articles
Journal of Robotic Systems - Special Issue on the DARPA Grand Challenge, Part 2
Multi robot mapping using force field simulation: Research Articles
Journal of Field Robotics
PerMIS '07 Proceedings of the 2007 Workshop on Performance Metrics for Intelligent Systems
DGCI'05 Proceedings of the 12th international conference on Discrete Geometry for Computer Imagery
P-SLAM: Simultaneous Localization and Mapping With Environmental-Structure Prediction
IEEE Transactions on Robotics
Hi-index | 0.00 |
We present a concept and implementation of a system to integrate low level and mid level spatial cognition processes for an application in robot mapping. Feedback between the two processes helps to improve performance of the recognition task, in our example the alignment of laser scans. The low level laser range scan data ('real scans'), are analyzed with respect to mid level geometric structures. The analysis leads to generation of hypotheses (Virtual Scans) about existing real world objects. These hypotheses are used to augment the real scan data. The core mapping process, called Force Field Simulation, iteratively aligns the augmented data set which then in turn is re analyzed to confirm, modify, or discard the hypotheses in each iteration. Experiments with scan data from a Rescue Robot Scenario show the applicability and advantages of the approach.