Date of final exam: 25/02/2005

E-Mail: dlongo@diees.unict.it

Home page: www.robotic.diees.unict.it

Tutor: Prof.  G. Muscato, Università degli Studi di Catania

  _______________________________________________________________________________________________________________________

Climbing robots: applications, design methodologies, control, experimental results

  _______________________________________________________________________________________________________________________

Award: 15 February 2008 - “Bando di concorso a premi di studio per giovani studiosi” supported by Gioenia Academy of Catania, in "SVILUPPI ED APPLICAZIONI DELLA ROBOTICA"

Advisor: Prof.  G. Muscato, Università degli Studi di Catania

Summary of the thesis

A very important problem in many industrial environments is the periodic inspection of the plants. The purpose of these inspections is to prevent dangerous plant operations due to structural problems. For example crack or corrosion on a petrochemical tank can led to leak with considerable ambient damage and risk for people living or working around the area. These inspections have to be performed at regular rate as standardized by the respective controlling body. All the inspection procedure implies that the plant must be put out of service and must be decontaminated to allow specialized staff to perform their work. Nevertheless the procedure itself can be very dangerous for technicians and is very costly. A possible solution can be to perform all the requested inspection procedures by using an automatic vector (climbing robot) that can climb the wall, whatever the material is made of, and use suitable sensors for the current inspection. The aim of this work is to present several design and control methodologies exploited in the development of some climbing machines for inspection purpose. In the first two chapters, current inspection methodologies and related problems will be analyzed. In this phase some key points that a climbing machine has to respect, will be pointed out. Mainly it has to be very light and to use very simple kinematics; these principles led generally to a cheaper machine. Moreover existing climbing machines will be also analyzed. In order to meet these requirements a new mechanism to climb walls based on sliding locomotion will be presented. Starting from the basic idea, in chapter three, four and five three prototypes and the related design methodologies will be illustrated. All these machines were tested in different conditions and optimized also by using soft-computing techniques. In the last part some relevant test results of the system in typical working conditions are presented. Parts of this work was supported by the Murst Project ISR1 “Sviluppo di metodologie innovative per la previsione, la mitigazione ed il controllo dei rischi derivanti da attività industriali chimiche e petrolifere” and by the Clawar I and Clawar II thematic network on Climbing and Walking Robot that are two project founded by the European Community.

_______________________________________