Multi-Sensor Fusion Method For Mobile System Localization
Proceeding: The Fourth International Conference on Artificial Intelligence and Pattern Recognition (AIPR)Publication Date: 2017-09-18
Authors : Wassila Meddeber; Youcef Touati; Arab Ali-Cherif;
Page : 17-20
Keywords : Localization; Mobile Robotics; Data Fusion; Multi-sensors; Bayesian Filter; Kalman Filter; Particle Filter; Smart Wheelchair;
Abstract
This paper discuss the multi-sensor data fusion problem of localization system for mobile robot of the wheelchair type, to help disabled and elderly people. A Kalman-Particle Kernel Filter (KPKF) method is proposed. This method is based on a hybrid Bayesian filter, combining the extended Kalman filter and the particle filter. Indeed, the KPKF models the conditional density at each instant by a Gaussian mixture in which each component has a small covariance matrix (density estimate). The Kalman correction is added to the weight correction to bring the particles back into the most probable space area. This method is applied in non-linearity and multimodality environment in order to improve the localization performance and the constraints encountered in previous filters. The implementation of our approach is carried out on an experimental platform of LIASD Smart Wheelchair navigating in a known indoor space.
Other Latest Articles
- Combined Neural Network Model for Real Estate Market Range Value Estimation
- Mapping of Relational Schema to Ontology Model
- DEMENTIA: FROM SUSPICION TO TREATMENT
- ACTUAL PROBLEMS OF INSTITUTE OF PETITIONS AND COMPLAINTS AT THE INVESTIGATIVE STAGE OF A CRIMINAL CASE
- MORPHOLOGY OF COMPENSATORY AND ADAPTIVE REACTIONS IN THE PLACENTA OF MULTIPARA AT AN UNCOMPLICATED PREGNANCY IN REPUBLIC OF UZBEKISTAN
Last modified: 2017-10-02 23:39:34