logo SBA

ETD

Archivio digitale delle tesi discusse presso l’Università di Pisa

Tesi etd-09022020-164723


Tipo di tesi
Tesi di laurea magistrale
Autore
DONATO, ENRICO
URN
etd-09022020-164723
Titolo
Underwater Mobile Manipulation with a bio-inspired legged robot
Dipartimento
INGEGNERIA DELL'INFORMAZIONE
Corso di studi
BIONICS ENGINEERING
Relatori
relatore Prof. Calisti, Marcello
Parole chiave
  • bio-inspired
  • force ellipsoid
  • gough-stewart platform
  • inverse kinematics
  • kinematic controller
  • legged robot
  • manipulability ellipsoid
  • parallel robot
  • path following
  • teleoperation
  • underwater
Data inizio appello
09/10/2020
Consultabilità
Non consultabile
Data di rilascio
09/10/2090
Riassunto
SILVER2 is an underwater legged robot employed to detect micro plastic elements into the sediment, and to clean the seabed from litter. Within the context of actual field missions, the goal of my thesis is to control the legs of the robot while performing specific tasks such as station keeping and manipulation.
The robot has been treated as a parallel robot, a mechanical system that uses more than one serial kinematic chain to support a single platform, or end-effector. By taking as a reference the control of the Gough-Stewart platform, the robot is able to assume admissible postures, placing constraints both on position and orientation of the body. Different algorithms have been developed to reach goal configurations and to perform path following, with the overall objectives of performing both teleoperated and autonomous tasks. The algorithms have been implemented in a kinematic simulator, integrated into the robotic operating system, and employed during actual sea trials for position withstanding and litter collection. The comparison between the kinematic simulator and the robot postures, altogether with the success of the sea operation, demonstrated the effectiveness of the algorithms. Eventually, the kinematic controller was complemented with simulations of force and manipulability ellipsoids, with the goal of maximizing forces along desired directions during manipulation tasks.
File