Please use this identifier to cite or link to this item: `http://hdl.handle.net/11455/74703`
 標題: 有限字元考量下應用Delta運算子設計數位卡曼濾波器之研析 作者: 章賢聰 關鍵字: δ運算子;q運算子;量化誤差;數位卡曼濾波器 出版社: 國立中興大學工學院;Airiti Press Inc. Project: 興大工程學刊, Volume 14, Issue 3, Page(s) 155-169. 摘要: The discrete system, obtained by sampling a continuous linear system for the conventional discrediting methods (q operator), could be out of its design performance as the sampling frequency is increasing. On the other hand, when the discrete system is implemented on hardware or on computer simulation, the ideal design performance could also be changed because of the quantization error effects on the coefficients and states. Therefore, in this paper, a δ operator is adopted as an alternate discretizing method for the reasons that its discrete mathematical model converges to the continuous counterpart as the sampling frequency is increasing to infinite and it could be more accuracy in numerical calculation. Together with δ operator and mathematical model of quantization error, t he main purposes in this paper are to analyze the quantization effects on digital system and discrete Kalman filter.以傳統數位方法(q運算子)所獲得之數位系統，在取樣頻率增加至非常大時，可能會產生失真的現象，無法趨近於理想的連續系統。此外，數位系統在實體化的過程中，亦會因系統之系數及狀態受到有限字元量化的影響，而使系統性能產生變化，無法達到理想的設計規格。基於此，本研究中，引用δ運算子做為數位化的方法，以其在數學模式表徵方式上之一致性，及在數值計算模擬中具有較為精確之優點，結合量化誤差之數學模式，對於數位系統及數位卡曼濾波器等主題，進行量化誤差分析。 URI: http://hdl.handle.net/11455/74703 ISSN: 1017-4397 Appears in Collections: 第14卷 第3期工學院