An Optimal Design of Target Tracker by α-β-γ-δ Filter with Genetic Algorithm

應用α-β-γ-δ過濾器及基因演算法於目標追蹤器最佳化設計

吳村木(Chun-Mu Wu) ； 張清靠(Ching-Kao Chang) ； 朱東德(Tung-Te Chu)

《中國機械工程學刊》 30卷6期 (2009/12) Pp. 467-474

https://doi.org/10.29979/JCSME.200912.0003

摘要

本文是使用α-β-γ-δ過濾器、基因演算及數值模擬，探討追蹤目標的方法。α-β-γ過濾器，已應用於追蹤目標及預測問題；本文使用新的α-β-γ-δ過濾器研究，結果顯示，精確度有顯著的改善，但有一缺點，在於運算過程，需用較多的電腦計算時間，若應用基因演算法（稱之基因α-β-γ-δ過濾器），可以降低時間，同時，本文使用數值模擬分析，確實求得最佳化的結果。

並列摘要

This study is concerned with a target tracker by α-β-γ-δ filter with genetic algorithm (GA) under numerical simulation. Generally, the third-order tracker of α-β-γ filter solution is presented for the target tracking and predicting problem for years. The filter can track the target's position and velocity, but not the acceleration. In order to improve the tracking accuracy, a new fourth-order target tracker called α-β-γ-δ filter can be explored. As a result, it exhibits significant improvement in tracking accuracy over the α-β-γ filter. Nevertheless, the only weakness of α-β-γ-δ filter is to use much computational time in optimization process. Therefore, a GA optimization of the parameters for α-β-γ-δ is proposed, which the new target tracker is called GA-based α-β-γ-δ filter and it is to reduce in computational time substantially. Meanwhile, the simulation is included to illustrate the analysis.

並列關鍵字

target tracker ； genetic algorithm ； α-β-γ-δ filter ； GA-based α-β-γ-δ filter