Adaptivepd- Controller For Positioning Of Rov

ARTICLE IN PRESS

Control Engineering Practice 15 (2007) 411–419 www.elsevier.com/locate/conengprac

Adaptive PD-controller for positioning of a remotely operated vehicle close to an underwater structure: Theory and experiments
Nguyen Quang HoangÃ, Edwin Kreuzer
Hamburg University of Technology, Mechanics and Ocean Engineering, Eissendorfer Strasse 42, 21073 Hamburg, Germany Received 24May 2005; accepted 4 August 2006 Available online 7 November 2006

Abstract The requirement for high accuracy in dynamic positioning of remotely operated vehicles (ROV), especially when tasks close to underwater structures have to be performed, demands high precision of sensor systems. Taut-wire and passive arm systems can satisfy this demand in measuring ROVs positions and orientations relativeto a structure. However, the main drawback of these sensor systems is that additional forces act on ROVs due to the mechanical connection. In order to solve this problem, an adaptive PD controller is proposed and designed for dynamic positioning of ROVs working in close proximity of structures. Invoking the adaptation law, these additional forces caused by the passive arm and umbilical, and evenby the uncertainties in gravity and buoyancy can be identified and compensated. By choosing an adequate Lyapunov candidate function, the system’s stability is proven. The effectiveness of this design control method is demonstrated by means of numerical simulations and experiments. r 2006 Elsevier Ltd. All rights reserved.
Keywords: Remotely operated vehicles (ROVs); Underwater vehicles; AdaptivePD control; Dynamic positioning

1. Introduction The application of a ROV is mainly divided into three phases. Firstly, the vehicle is launched into water from the mother-ship and is then controlled to reach a structure. Secondly, the vehicle conducts assigned tasks, e.g. inspection or maintenance around the structure. In the last phase, the vehicle is navigated back to the mother-ship (Pinto,1996). The Petri net-based approach introduced in Caccia, Coletta, Bruzzone, and Veruggio (2005) can be used to switch the operation of the ROV from one phase to another phase. The technical features of sensors may vary according to the working phases. Consequently, the vehicle should be equipped with different sensor systems for the different performance phases. For the first phase, travelingfrom mother-ship to the underwater structure, the required precision of the sensor system may be low. Therefore, a navigation system based on wave-propagation or the
ÃCorresponding author. Tel.: +49 40 4 2878 2355; fax: +49 40 4 2878 2028. E-mail address: nguyen.hoang@tu-harburg.de (N.Q. Hoang).

inertial navigation system is usually used to determine the position of the vehicle. In the secondphase, working close to the structure, the relative position of the ROV with respect to the structure is important in positioning the ROV precisely. Accordingly, the requirement for the accuracy of the sensor system is significantly higher than that in the first phase. For the purpose of this study only the second phase is considered. The navigation sensor system used in the first phase can still beapplied to this phase. However, many operations in the second stage require to maintain a ROV close to the structure. Moreover, inspecting a structure is usually the most time-consuming task throughout the ROVs working duration. Thus, a sensor system with a mechanical connection to the structure is suggested because it can measure the ROVs positions more accurately than those done by thenavigation systems (Cunha, Scieszko, Costa, Hsu, & Anna, 1993). So far, taut-wire and passive arm are two of such a mechanically connected sensor system. Taut-wire is a cable stretching between a ROV and its underwater structure. The relative position of the ROV is calculated by measuring the cable length and two angles.

0967-0661/$ - see front matter r 2006 Elsevier Ltd. All rights reserved....