Año: 2017
Autores: Gimenez, J.; Tosetti, S.; Salinas, L.; Carelli, R.;
Resumen: This work presents a non parametric probabilistic mapping based on kernel estimators which does not use grids.The proposed methodology characterizes the map with a cloud of points obtained from several observations of the environment. In order to maintain a bounded number of observations in memory, a recursive subsampling algorithm is proposed. The procedure is included in an SLAM, in order to localize the robot as well. An application example is the presented, where the proposed methodology is applied in an agricultural environment. Simulation and experimentation results are presented to validate the proposal.
Link: https://ieeexplore.ieee.org/abstract/document/8214350