Leif Christensen, DFKI GmbH, Bremen/Germany, firstname.lastname@example.org Niklas Fischer, BALance TC GmbH, Bremen/Germany, email@example.com Sven Kroffke, DFKI GmbH, Bremen/Germany, firstname.lastname@example.org Johannes Lemburg, DFKI GmbH, Bremen/Germany, email@example.com Reinhard Ahlers, BALance TC GmbH, Bremen/Germany,firstname.lastname@example.org Abstract In this paper we describe the technical concept and first implementation of an autonomous rail guided robot for inspection and maintenance tasks in ballast water tanks (BWTs), especially for double bottom BWTs. We evaluate different locomotion techniques and other robotic principles potentially suited for such application scenarios and propose a modular concept that uses asimple corrosion-free thermoplastic rail which runs 'through' the robot and thus allowing for high payloads and upside-down operation. As a second aspect the paper describes the assessment of life cycle benefits for the developed system with cost parameters like scaffolding costs, tank preparation-, intank inspection- and vessel out-of-service time. 1. Introduction Ballast water is used tostabilize partially loaded or empty ships on the open sea. When needed, ocean or port water is pumped into special ballast water tanks to increase the mass of a ship. Because ocean water is typically contaminated with algae, plankton and other organisms, and due to the aggressive nature of salt water, BWTs are often subject to serious fouling and corrosion. The ballast tanks represent more than 40% ofthe entire coated area on a vessel. They therefore need frequent inspection, cleaning, and repair. Until now, the maintenance of BWTs cannot be performed under operating conditions. Consequently, ship owners are forced to send their ships to dock inspection every 2.5 or 5 years, causing high costs in labour and ship-downtime. On cruise vessels BWTs are typically built in those spaces of a ship thatcannot be used for other purposes. As a consequence, they are narrow, irregularly shaped, and badly ventilated. They are designed to be accessible to human workers, but are surely a dirty, unhealthy, and unpleasant workplace, Fig. 1 and Fig. 2.
Fig. 1: Double-bottom BWT of a cruise liner at Meyer Werft Shipyard (construction phase) Manual work in a BWT is a tedious task that carries potentialshort- and long-term health risks for the workers involved. Nevertheless, coating, cleaning, inspection and repair of ballast water tanks is still done manually. An automation of these processes is not feasible so-far due to the complexity and the
variability of current tank-design, as well as the limited flexibility of the available robotic systems. If robots were to take over theinspection, cleaning, coating, and repair of ballast water and other tanks and narrow spaces, this would greatly speed up the process and reduce the health risks for workers and ship crews. Using robots, ship builders, ship owners and class societies alike would be able to considerably reduce the cost for inspection and maintenance of BWTs. The international R&D project ROT (RObots in Tanks) addressesthese challenges by evaluating available robotic principles potentially suited for BWT inspection (described in section 2), developing a concept for a BWT inspection robot and implement the concept in a robotic demonstrator (section 3). The whole development is accompanied in the project with a life cycle cost analysis of such a robotic system and the first findings on that are shown in section 4.Section 5 concludes the work done so far and gives an outlook for next possible steps. 2. General robotic concepts potentially suited for BWT inspection Although the inspection and maintenance of BWTs today is done manually, there are generic robotic construction principles, systems and system components, which have to be evaluated to find a starting point for a design of a robot potentially...