| Summary: | 碩士 === 逢甲大學 === 自動控制工程所 === 92 === The purpose of this research is to accomplish the key techniques：real-time auto-tuning of the motion-cue parameter and hardware-in-the-loop simulation. This constructed tank can satisfy the training requirements of pilot-in-the-loop simulation in tank, which is the focus of this research. For the motion-cue algorithm, coordinate genetic algorithm, coordinate adaptive fuzzy washout algorithm and digital control washout algorithm will be proposed in this study to achieve the requirement of high motion-cueing fidelity with in the limited platform working space. As we know, all the parameters of the classical washout algorithm must be generated by hand-tuning process in the thesis. The optimal parametes of the motion cueing can be searched globally and validated automatically under some constrained conditions by the coordinate genetic algorithm. Using coordinate adaptive fuzzy washout algorithm, we can auto-tune the motion cueing parameters of the tank simulation according to the varying frequency of the dynamic input signal in real-time. Therefore this method has possessed the capability of real-time tuning. Digital control washout algorithm is only suitable to the specific input acceleration signal. For the construction of hardware-in-the-loop, we accomplish the real-time interactive remote control programs by transfening the position and attitude of tank to the monitor program through the internet, and real-time react to the behavior of the tank according to the virtual reality scene.
|
|---|
