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
Object modelling by registration of multiple range images
Image and Vision Computing - Special issue: range image understanding
Iterative point matching for registration of free-form curves and surfaces
International Journal of Computer Vision
The map-building and exploration strategies of a simple sonar-equipped mobile robot
The map-building and exploration strategies of a simple sonar-equipped mobile robot
Directed Sonar Sensing for Mobile Robot Navigation
Directed Sonar Sensing for Mobile Robot Navigation
Robot Pose Estimation in Unknown Environments by Matching 2D Range Scans
Journal of Intelligent and Robotic Systems
Occupancy grids: a probabilistic framework for robot perception and navigation
Occupancy grids: a probabilistic framework for robot perception and navigation
Directional Processing of Ultrasonic Arc Maps and its Comparison with Existing Techniques
International Journal of Robotics Research
Metric-based iterative closest point scan matching for sensor displacement estimation
IEEE Transactions on Robotics
Visual Localization Using Ground Points
Proceedings of the 2010 conference on Artificial Intelligence Research and Development: Proceedings of the 13th International Conference of the Catalan Association for Artificial Intelligence
| Hi-index | 0.00 |
Scan matching algorithms have been extensively used in the last years to perform mobile robot localization. Although these algorithms require dense and accurate sets of readings with which to work, such as the ones provided by laser range finders, different studies have shown that scan matching localization is also possible with sonar sensors. Both sonar and laser scan matching algorithms are usually based on the ideas introduced in the ICP (Iterative Closest Point) approach. In this paper a different approach to scan matching, the Likelihood Field based approach, is presented. Three scan matching algorithms based on this concept, the non filtered sNDT (sonar Normal Distributions Transform), the filtered sNDT and the LF/SoG (Likelihood Field/Sum of Gaussians), are introduced and analyzed. These algorithms are experimentally evaluated and compared to previously existing ICP-based algorithms. The obtained results suggest that the Likelihood Field based approach compares favorably with algorithms from the ICP family in terms of robustness and accuracy. The convergence speed, as well as the time requirements, are also experimentally evaluated and discussed.