5.1 The Full Mission Bridge Simulator (EMBS) in KRISO/KORDI

　

　

 We test the performance of VCGS using the 3D FMBS in KRISO/KORDI. Fig.5 shows the bridge of the FMBS where we set up the VCGS console with software of interface, sound cards, microphones and speaker. Therefore, we integrate all navigational equipment with VCGS like IBS, using the network. In addition, to raise the rate of SR we tune up the environmental noise in bridge through the optimization of three tuning values; volume of microphone, minimum range of voice amplitude, and the detection factor of erroneous pronunciations.

　

5.2 The Real-time Simulation of VCGS

　

(1) Scenarios of simulation

　

 We assume the situation of arrival in Busan Harbor and repeatedly test 3 scenarios of simulations according to VCGS, wind, current, and visibility.

　

　

(2) The Passage Planning of Simulation

　

 A operator, who has is the navigational officer tests the real-time simulation; i.e. OPBO.

　

　

 The ship starts far from Harbor limit and docks at the port of container. Auto-pilot is used until the ship arrives near to the breakwater of outer port. From then on, the manual steer is used.

　

 Passing ships (containers) that depart from the inner port will encounter the own ship near the breakwater of the outer port.

　

 VCGS starts from Stand-by mode and changes to Scheduled Report (SR), Emergency Report (ER), or Query Report (QR). We set the time interval of SR to 5 minutes. In addition, we set up the SR for harbor information to execute each time when the ship passes near appropriate objects.

　

5.3 Results of real-time simulation

　

 Fig.6 shows the example (001.ARRIVAL.001) out of results of simulation. We illustrate the trajectory at 30 seconds and display the ship in red and passing ships in black.

　

 To evaluate the result of simulation, we use just qualitative measurement, that is, subjective evaluation of the navigational officer that tested the simulation. The navigational officer evaluates the simulation in respects of familiarity, convenience, and accuracy.

　

 Table 3 shows the evaluation results for functional performance of VCGS and each evaluation value is the average of all results in all scenarios (5: very satisfactory, 4: satisfactory, 3: normal, 2: unsatisfactory, 1: very unsatisfactory).

　

　

 From Table 3, we get the satisfactory evaluation for VCGS. Consequently, using VCGS, the navigational officer quickly grasped the navigational information and comfortably controlled the ship. That is to say, owing to user-friendly concepts and functions, user familiarly used VCGS, and rapidly grasped the navigational condition of the ship.

　

 After the simulations, we get some comments for function of VCGS from navigational officer. He suggests that VCGS should offer the ways of collision avoidance or grounding prevention in emergency. Maybe, if the expert system that contains the prevention of emergency develops, we will be able to solve this problem.

　

 On the other hands, we need to find a way of accurately acquiring the position and velocity of other ship, because they are very important information to analyze the collision risk in emergency. To be sure, using the AIS, we can get accurate information about the position and velocity of other ship. However, It takes time of limit specified in the law to set up the AIS to all ships that are currently navigating.

　

 We design the user-friendly VCGS to make VCGS more practicable for safe navigation. By using interviews, we survey functions and procedures that navigation officers want to be included in VCGS. Moreover, we apply a user-independent SR engine so that a procedure of learning of speakers is not necessary.

　

 In order to reduce the error due to low rate of SR and make navigational officer conveniently use VCGS, we apply the user-friendly automation and functional modes for self-correction in procedures of report. In addition, to grasp navigational condition of the ship quickly, we integrate the interface of VCGS with CID.

　

 To evaluate the performance of VCGS, we carried out the real-time simulation using the 3D FMBS in KRISO/KORDI by the present navigational officer.

　

 From the real time simulations, we get the satisfactory evaluation for VCGS. Owing to user-friendly concepts and functions, user familiarly used VCGS, and rapidly grasped the navigational condition of the ship from VCGS that we integrated with CID. Therefore, the navigational officer comfortably controlled the ship and quickly copes with the emergency. Consequently, the operator can reduce the fatigue of work in bridge using VCGS more largely than in manual procedure, so the users can simultaneously carry out other work (multitasking) with a lookout. That is to say, navigational officer can carry out more safe navigation using VCGS.

　

 In this research, because of the problem of test place, we apply VCGS to FMB simulator instead of the real bridge like IBS. In the future, we need to set up this VCGS in the real bridge like IBS and test the performance in the condition of real sea.

　

[1] N.S.Son, S.Y.Kim, "Development of a supporting system for safe navigation using Voice Control and Guidance", Journal of Ships and Ocean Engineering, Vol 33, p. 117-126,2002

[2] Text-To-Speech and Speech Recognition System, Dong-yuk Mechatronics Institute, Young-Jin Publishing Co., 1993

[3] The present status and prospects about technologies of Speech Recognition, Proceedings of Electronic Engineers of Korea, Vol 20, No.5,pp.548-556,1993

[4] Junji Fukuto, Yasuyoshi Itoh and Masayoshi Numano, "Evaluation of Integrated Navigation System with Speech Communication", Journal of Japan Institute of Navigation, No. 136, pp.26-35

[5] Kazuo Matsuda, "Navigational Aid by Voice Control and Guidance", Journal of Japan Institute of Navigation, No. 130, pp. 17-23, 1996

[6] C. A. Kamm & M. A. Waker, "Design and Evaluation of Spoken Dialogue Systems", 1997 IEEE Workshop on Automatic Speech Recognition and Understanding Proceedings,pp. 11-18,1997

[7] Marie Meteer, "Voice Input for Collaborative Systems", 1997 Workshop on Automatic Speech Recognition And Understanding Proceedings, pp. 19-25,1997

[8] N.S.Son, S.Y.Kim, "Development of Voice Information System for Safe Navigation in Marine Simulator", Journal of the Korean Society for Marine Environmental Engineering, Vol.5, No.3, pp.28-34,2002

[9]H.J.lee, "Optimization of Collision Avoidance System", a thesis for a doctorate, Seoul National University, pp.8-10, 1997

[10] Bridge Procedures Guide, International Chamber of Shipping, Witherby & Co.,Ltd, 3^rd edition, 1998

　

 This research is supported by two research projects, "Development of Base Technology for Integrated Maritime Risk Management System" and "Development of Automatic Control System of Ship using Voice Technology" of KRISO/KORDI.

　

 N.S.Son is a researcher of Korea Research Institute of Ships and Ocean Engineering (KRISO/KORDI) in Korea.

　

 S.Y.Kim is a team manager and senior researcher of Korea Research Institute of Ships and Ocean Engineering (KRISO/KORDI) in Korea.