"本研究計畫擬提出粒子群演算法求解雙天體天文船位問題，進而延伸求解引進截距法概念的過定天文 船位問題。從數學觀點，過定天文船位的合理解集合應是天文船位及其誤差範圍；又由於觀測天體高 度必存在誤差，故提供過定天文定位之散佈圖的預判機制以符合實務作業。值得注意的是對不同的觀 測時間，以航進定位概念並採用中緯航法移動參考點，藉此計算其在定位時間的天文船位。本研究計 畫擬以粒子群演算法自行開發程式，命名為nAVPPro_PSO，以提供航海者使用。最後，將挑選數個標 準計算實例展示成果並驗證本研究方法的準確性與其廣泛的適用性。""In this proposal, a particle swarm optimization (PSO) algorithm will be developed to deal with the two-body astronomical vessel position (AVP) problem and is further extended to resolve the overdetermined AVP (nAVP) problems by combining with concept of the intercept method. From viewpoint of mathematics, the reasonable solution should fall into the real AVP within an acceptable error range. When the sight is conducted in practice, various errors will produce. Therefore, a prejudge mechanism should be offered in the scatter chart. It is noted that when the non-simultaneous sights condition is encountered, the running fix concept and the mid-latitude sailing will be adopted to translate reference points to the fix time for determining the AVP. Based on the PSO, a program namely nAVPPro-PSO will be developed for the navigator. Finally, several validation examples will be provided to show the accuracy and versatility of the PSO."
particle swarm optimization
the overdetermined astronomical vessel position