3D underwater SLAM using sonar and laser sensors

This thesis presents a SLAM algorithm using a multibeam echosounder, a new 1/2 underwater laser scanner for online 3D perception and two applications of this new laser scanner: inspection and intervention. The SLAM algorithm using a multibeam echosounder presents a new 3D probabilistic registration method. The results of this registration method are used as updates of an extended Kalman filter where the positions of the different submaps is estimated. This thesis also presents the model and calibration of the new underwater laser scanner. This new sensor outperforms in resolution, precision and 3D scanning frquency the acoustic multibeam and the state of the art laser scanners. Finally, this sensor in used in the inspection, using SLAM, of a submerged structure, and in two manipulation experiments. One using path planning in an unknown environment and the other one grasping an object at the bottom of the water tank ​
​L'accés als continguts d'aquesta tesi queda condicionat a l'acceptació de les condicions d'ús establertes per la següent llicència Creative Commons: http://creativecommons.org/licenses/by/4.0/