Paloma de la Puente
Menu:
Research
2D maps
I have implemented a Maximum Incremental Probability Algorithm based upon the Extended Kalman Filter for localization and map building and combined it with a novel method to remove dynamic points from the generated maps. Integration with reactive and motion control allowed for the development of a simple yet effective navigation system for mobile robots.
Here are some of the results I obtained when testing the algorithms with both simulated and real data. For more details see the Publications section.
Interior of the Intel Research Lab in Seattle
| Map built upon odometry data solely | Map built with the above mentioned algorithm |
 |
 |
This data set was obtained from the Robotics Data Set Repository (Radish). Thanks go to Dieter Fox for providing this data
Interior of the Principe Felipe Science Museum
within the City of Arts and
Sciences in Valencia, Spain
Office Environment for MobileSim simulator by MobileRobots.Inc
 |
 |
Lab
Corridor
3D maps
Moving on to the third dimension, I have developed a plane-based Maximum Probability Algorithm for localization and map building.
Here are some of the results I obtained when testing the algorithms with real data acquired at our laboratory by our robot Nemo. For further details see the Publications section.
Lab
Detailed scenes
 |
 |
Segmentation process
 |
 |