After submarine wreck, deep submergence rescue vehicle is the important method for submariners escape. Especially, after the incident of Russia's "Kursk" sank in recent years, states pay more attention to the work of Submarine Rescue, and increase research and development of deep submergence rescue vehicle.In this paper, we design the overall structure of dynamic positioning system according to the requirement of deep submergence rescue vehicle dynamic positioning system. This paper achieved a deep submergence rescue vehicle information display system, which based Windows operating system platform, using Visual C ++ 6.0 IDE. Based on the VxWorks operating system platform, this paper divised the tasks for dynamic positioning system control functions and developed a real-time control system using the multi-tasking mechanism.Upon in-depth analysis of requirement of the dynamic positioning system of deep submergence rescue vehicle, this paper designed a multi-modal RBF neural network self-tuning control algorithm, which is achived on the VxWorks operating system platforms. Semi-physical simulation studies have shown that dynamic positioning system designed by this paper has easy operability, friendly man-machine interface and real-time control. At the same time, the simulation results also show that the proposed multi-modal RBF neural network self-tuning control algorithm is effective and practical in the ocean currents, especially in changing ocean currents. This research will be used in the product design of deep submergence rescue vehicle dynamic positioning system.
|