Submarine Automation : Demonstration # 5

This paper describes the results obtained in the final performing period of the ARPA sponsored submarine automation project1. Efforts on the mapping between the submarine operational environment and the RCS software architecture lead to the result of three watch station graphic user interface panels. The submarine automation model has been expanded to include some engineering systems control capability. On the RCS generic structure, authors have explored methods to reuse existing plans for new task requirements. The authors have also illustrated a main feature in RCS, namely, a smooth transition of level of authority in commands from the higher to the lower control levels. 1.BACKGROUND AND PREVIOUS WORK An earlier paper [Hu 93-1] describes the following: submarines are complex systems. Navigation, communication, hydrodynamic control, power, etc., are just a few among all the subsystems that need to be coordinated when submarines conduct missions. An enormous amount of information must be fused, organized, and communicated to support decision making in real-time. On today’s submarines, most of these functions are performed by crew members in extremely tight space. It is, therefore, very desirable to have submarine operations automated. The Intelligent Systems Division (ISD) of the National Institute of Standards and Technology (NIST) has been supporting the Advanced Research Projects Agency (ARPA) Maritime Systems Technology Office (MSTO) in investigating submarine automation. Our previous accomplishments include a demonstration of an automated maneuvering system for a 637 class nuclear submarine to perform under-ice transits in the Arctic region. The control system is capable of maneuvering the simulated submarine toward its intended destination while using simulated sonar data to avoid dangerous ice keels and to maintain the submarine's ordered depth. The control system can operate either autonomously or 1 ARPA Order No. 7829, Amendment No. 02.