仿生水下載具使用動壓力迴授追隨固體邊界或振動源之導航方法

Abstract

在自然界中，魚類可以藉由偵測周遭水域的水動壓變化來獲得鄰近的訊息，這個概念亦適用於仿照魚類游動而設計的仿生型水下載具，本研究開發了藉由回饋水動壓變化以控制魚型載具的導航方法。為了控制一台機器魚於水中沿著固體邊界游動，從而對周遭環境進行探索，本文提出了一種藉由壓力回饋控制機器魚沿著直壁游動的方法。此方法基於勢流理論，其中機器魚的尾部和壁面效應分別被描述為一偶極子、以及在牆壁的相對側上的鏡像偶極子。從這個模型中，機器魚相對於牆壁的位置可由裝配在機器魚頭部和身體上的壓力感測器所測得之壓力數據的大小和比值導出，而這些得到的訊息將用於控制機器魚的方向。實驗結果呈現機器魚可以被控制在距離牆壁(間隙/翼展)比為1-1.33內沿著游動，這也顯示出以水動壓迴授控制機器魚追隨固體邊界的可行性。 此外，為了達成群游，機器魚必須能偵測鄰近個體的周期性運動，並配合調整自身運動。為了達成此目標，本研究另提出一控制機器魚的尾部運動以跟隨鄰近振源的振動的方法。機器魚的尾部運動和振源的振動被描述為兩個振動器，並可藉由同步振動器的方法來達成相位跟隨。為了追踪振動源，機器魚周圍的流體動壓是由PVDF壓電感測器來量測。從測量到的壓力中減去機器魚產生的壓力預測值，即可獲得由振動源產生的壓力。同樣利用建置於勢流理論的偶極子模型，機器魚尾部和振動源之間的動作相位差，即可自PVDF感測器量測到的壓力值推估，並將用來決定驅動尾部的轉矩大小。自實驗中以被局部限制的機器魚進行跟隨鄰近振動源的控制，結果顯示本文所提出的以迴授水動壓以進行相位跟隨控制方法是有效的。

In nature, fish can extract near field information via detecting nearby pressure variations. This concept is suitable for a biomimetic autonomous underwater vehicle (BAUV): a fish-like swimming robot. This study develops the methods to control the robotic fish by feeding back the hydrodynamic pressure variations of the environment. To control a robotic fish to swim along the solid boundary while exploring the environment, this paper presents a method to control the robot to swim along a straight wall via pressure feedback. A model based on two-dimensional potential flow is used, where the tail of the robotic fish and the wall effect are described as a dipole and an image dipole on the opposite side of the wall, respectively. From this model, the position of the robotic fish with respect to the wall can be derived from the magnitudes and ratios of the pressure data measured by the pressure sensors equipped on the head and the body of the robot. The information is then used to control the direction of the robotic fish. From the experiments, the robotic fish can swim a distance-to-wall/tail-span ratio of 1-1.33 beside a wall, which shows the effect of the proposed method. Besides, a robotic fish is expected to detect the periodic movements of its neighbors and adjust its motion, such that it can swim properly in a school. To achieve that, this study also presents a method to control the tail motion of a robotic fish to follow the oscillation of a neighboring source. The tail motion of the robot and the oscillation of the source are described as two oscillators, and can be synchronized by an oscillator-based method. To track the oscillation of the source, the hydrodynamic pressure around the robotic fish is measured by a PVDF sensor on the robot. The predicted pressure generated by the robotic fish is subtracted from the measured pressure, and the pressure generated by the source is obtained. The pressure field is also described by a dipole model based on the potential flow theory. Based on this model, the phase difference between the tail and the oscillating source can be estimated from the pressure measurements of the PVDF sensor. This phase difference is then used to determine the torque which drives the tail. From the experiments that involve a captured robotic fish swimming close to an oscillating source, the proposed phase following control method is demonstrated to be effective.