Digital image processing (2nd ed.)
Digital image processing (2nd ed.)
The robot localization problem
WAFR Proceedings of the workshop on Algorithmic foundations of robotics
Bayesian classification (AutoClass): theory and results
Advances in knowledge discovery and data mining
Learning metric-topological maps for indoor mobile robot navigation
Artificial Intelligence
Localizing a Robot with Minimum Travel
SIAM Journal on Computing
Map learning and high-speed navigation in RHINO
Artificial intelligence and mobile robots
Navigating Mobile Robots: Systems and Techniques
Navigating Mobile Robots: Systems and Techniques
Mobile Robot Localization and Map Building: A Multisensor Fusion Approach
Mobile Robot Localization and Map Building: A Multisensor Fusion Approach
Lifelong Planning for Mobile Robots
Revised Papers from the International Seminar on Advances in Plan-Based Control of Robotic Agents,
Bootstrap learning for place recognition
Eighteenth national conference on Artificial intelligence
Mobile robot mapping and localization in non-static environments
AAAI'05 Proceedings of the 20th national conference on Artificial intelligence - Volume 3
| Hi-index | 0.00 |
Global localization is the problem of determining the position of a robot under global uncertainty. This problem can be divided in two phases: 1) from the sensor data (or sensor view), determine the set of locations where the robot can be; and 2) devise a strategy by which the robot can correctly eliminate all but the right location. The approach proposed in this paper is based on Markov localization. It applies the principal component method to get rotation invariant features for each location of the map, a Bayesian classification system to cluster the features, and polar correlations between the sensor view and the local map views to determine the locations where the robot can be. In order to solve efficiently the localization problem, as well as to consider the perceptual limitation of the sensors, the possible locations of the robot are restricted to be in a roadmap that keep the robot close to obstacles, and correlations between the possible local map views are pre-computed. The hypotheses are clustered and a greedy search determine the robot movements to reduce the number of clusters of hypotheses. This approach is tested using a simulated and a real mobile robot with promising results.