An introduction to variational methods for graphical models
Learning in graphical models
Robust Monte Carlo localization for mobile robots
Artificial Intelligence
Navigating Mobile Robots: Systems and Techniques
Navigating Mobile Robots: Systems and Techniques
Monte Carlo Localization with Mixture Proposal Distribution
Proceedings of the Seventeenth National Conference on Artificial Intelligence and Twelfth Conference on Innovative Applications of Artificial Intelligence
Robust global localization using clustered particle filtering
Eighteenth national conference on Artificial intelligence
FastSLAM: a factored solution to the simultaneous localization and mapping problem
Eighteenth national conference on Artificial intelligence
UAI'02 Proceedings of the Eighteenth conference on Uncertainty in artificial intelligence
Dynamic motion models in Monte Carlo Localization
Integrated Computer-Aided Engineering
| Hi-index | 0.00 |
Mobile robot localization is the problem of tracking a moving robot through an environment given inaccurate sensor data and knowledge of the robot's motion Monte Carlo Localization (MCL) is a popular probabilistic method of solving the localization problem By using a Bayesian formulation of the problem, the robot's belief is represented by a set of weighted samples and updated according to motion and sensor information One problem with MCL is that it requires a static map of the environment While it is robust to errors in the map, they necessarily make the results less accurate This article presents a method for updating the map dynamically during the process of localization, without requiring a severe increase in running time Ordinarily, if the environment changes, the map must be recreated with user input With the approach described here, it is possible for the robot to dynamically update the map without requiring user intervention or a significant amount of processing.