Simulation environment for the investigation of automatized cooperation of marine crafts

https://doi.org/10.1016/j.matcom.2008.02.024Get rights and content

Abstract

There are marine operations involving the cooperation of several marine crafts. For instance spill-over confinement, rescue, towing, etc. A simulation environment has been developed to analyse the cooperation problems that may arise along an operation, in order to develop procedures for better joint work of the ships. The final target is to add some automatization into the control of the ships, with on-board computers helping the ship captains to take the right decisions. The environment considers the ships as agents. The paper introduces the simulation environment and describes some examples concerning ship formation cases.

Introduction

There are marine operations, such as spill-over confinement, towing, rescue, etc., which involve the cooperation of several ships. In particular, the recent sinking of the “Prestige” oil tanker near the Spanish coast has shown some issues requiring further analysis. For instance, it was noticed that the usual operation for oil spill-over confinement consists in two ships towing a boom. Running some simulations in the environment described in this paper, put into light several coordination difficulties arising with just two ships towing the boom. More than two ships for this task cause still more difficulties. In such a case, the help of automatized real-time coordination procedures would be useful. This can be made with computers on board the ships, bringing advice to the ship captains. The purpose of our research is to add automatization to marine operations. The on-board computer interacts with the ship captain in a verbal form. All computers communicate and think about the operation evolution, to help captains to take optimal decisions for a good achievement of the general operation.

An important step for the research is to establish a simulation environment for the analysis of operations with several cooperating ships. The paper describes a first version of the environment. Some experiments about formations of ships have been already done, offering an interesting basis to analyse cooperation difficulties and possible solutions.

As reflected by the scientific literature, the research on ships working together has started recently. Several different scenarios have been considered. For instance about towing [5], about ship formation control [10], or about synchronization of two ships [6]. The general approach is to extend to the marine context a main stream of contemporary research about multirobot systems [9]. Given a mobile robot, its behaviour may be governed by a reactive [2] or by a deliberative [3] approach, or any combination of both. The term agent is used as an important metaphor, being applied in different areas with a variety of specific meanings. Robots taking own decisions can be considered as agents [7]. Several robots can work as a team, with proper strategies [1], [4], [8]. There are several issues to be considered to constitute a robotic team:

  • The existence or not of communication means between agents. In case of having communications, they can be more or less error free.

  • The type of coordination strategy. For instance, there is a central coordinator, or not. Decisions can be delegated in more or less extent. The coordination policy may change along an operation.

  • Team members can have different capabilities or dynamic characteristics: the team can be homogeneous or heterogeneous (perhaps with hierarchies).

  • The external conditions, which can change along an operation in the sea for instance.

The main approach of the simulation environment is that each ship is an agent. Each agent obeys to behaviour rules that are specified using the simulation facilities. The simulation environment is made with MATLAB–SIMULINK. The paper, after a description of the simulation environment, presents two simulation experiments about formation control, highlighting interesting phenomena.

Section snippets

Simulation environment design

According to [4] the cooperation of robots involves three main aspects. One is knowledge of one agent about other agents. Another is coordination between robots, which can be strong, light, or none. The last is organization, which can be strongly centralized, lightly centralized, or completely distributed. In this paper the following specifications have been considered:

  • The agents have knowledge about each other.

  • The agents are strongly coordinated. Each ship takes decisions taking into account

Details of the simulation environment

In the beginning of the research, it was decided to develop a flexible simulation platform, open to changes in several aspects of the simulation. Consequently, the simulation implementation is modular and prepared for incremental functional enrichment and improvement. The simulation environment has been developed with MATLAB and SIMULINK. The behaviour of the ships has been described with several local automata. The toolbox STATEFLOW has been used, and the results obtained have been

Examples of simulation studies

In the following two sections, two examples of simulation studies will be presented. Both examples have some specifications in common. Let us inform about it.

From years ago we conduct experiments in a towing tank facility near our University, in Madrid. The name of the facility is “Canal de Experiencias Hidrodinamicas de El Pardo (CEHIPAR)”. In particular, we use a 150 m × 30 m basin with a wavemaker. Five scaled down ships have been built for the research on cooperative marine operations. This is

Conclusions

In this paper a simulation environment has been introduced for the study of automatized cooperation in marine crafts operations. It is a modular and flexible environment where different cooperation protocols can be studied, and many types of operational scenarios can be defined. It is open for improvements and functional empowering. It can also be used for other robotic contexts, such aerial vehicles.

The simulation environment is being used to establish good cooperation protocol candidates, to

References (10)

  • E.A. Billard

    Patterns of agent interaction scenarios as use case maps

    IEEE Trans. Syst. Man Cybernet. Part B

    (2004)
  • R.A. Brooks

    A robust layered control system for a mobile robot

    IEEE J. Robot. Autom.

    (1986)
  • A. Elfes

    A distributed control architecture for an autonomous mobile robot

    Artif. Intell.

    (1986)
  • A. Farinelli et al.

    Multirobot systems: a classification focused on coordination

    IEEE Trans. Syst. Man Cybernet. Part B: Cybernet.

    (2004)
  • V. Johansen et al.

    Maneuvering of towed interconnected marine systems

There are more references available in the full text version of this article.

Cited by (4)

  • Influence of MCrAlY coating on low-cycle fatigue behavior of a directionally solidified nickel-based superalloy in hot corrosive environment

    2016, Materials Science and Engineering: A
    Citation Excerpt :

    Therefore, combined environmental and mechanical effects lead to premature failure by the surface formation of abundant sharp micro–cracks. Particularly, for the engines operated in marine environments, the ingested salt deposits, such as Na2SO4, NaCl, and V2O5, can develop on the surface of turbine blades [3–5]. These salts will lead to the severe hot corrosion of the turbine blades exposed to extremely high temperature [6–10].

  • Influence of the different salt deposits on the fatigue behavior of a directionally solidified nickel-based superalloy

    2016, International Journal of Fatigue
    Citation Excerpt :

    Combined corrosion and oxidation resistance, good mechanical performance, and microstructure stability at high temperatures [1] are among the desirable attributes of these superalloys [1,2]. In marine environments, salt from seawater, in combination with sulfur, vanadium, and other alkali metals present in the fuel, melt under the engine operating temperature, and induce hot corrosion of engine components [3–6]. Hot corrosion refers to the accelerated oxidation of materials at elevated temperatures induced by the fused salt deposit [7,8].

  • Preparing the automatic spill recovery by two unmanned boats towing a boom: Development with scale experiments

    2015, Ocean Engineering
    Citation Excerpt :

    First spill over experiments were presented in Leon et al. (2006) and Giron-Sierra et al. (2006). The path planning for the ASV team was considered in Giron-Sierra et al. (2008), and a simulation environment was presented in Carrillo et al. (2008). A general view of the main steps of the oil spill recovery operation was presented in Pereda et al. (2011).

  • The intelligent underwater laboratory

    2021, IOP Conference Series: Earth and Environmental Science
View full text