透過您的圖書館登入
IP:3.12.161.161
  • 期刊

Path Planning for the Automatic Pile Setting out

自動工程放樣之路徑規劃

摘要


本文使用類免疫演算法求取土木工程放樣路徑規劃之最佳路徑,因此可縮短放樣的工時,並減少土木建築工程的成本。放樣點增加時,其最佳路徑求解時間會大幅增加,在時間有限的情況下,是不可能以殆盡方式來求解。求取最佳路徑所需要的時間增加,因此無法在有限的時間內求出所有的路徑。類免疫演算法利用生物免疫系統裡,抗原與抗體之間的互動關係發展而成,免疫系統具有學習、聯想、記憶的特性,其中記憶的特性是多數演算法所沒有的。將最佳路徑視為抗原,而可行解當作是抗體,抗體經由不斷的演化,經過淘汰、複製、突變的機制,而使每代的抗體越趨優良,最終能夠在合理的時間內規劃出土木工程放樣路徑規劃的最佳路徑。

關鍵字

無資料

並列摘要


With the artificial immune system algorithm, this paper determines the optimal path planning for the pile setting out. Thus, we can shorten the work time of the setting out and the cost of civil construction will be reduced. But if the number of the setting out nodes increases, more time is needed to search the optimal path solution. And it is impossible to find out total paths in the limited time. Artificial immune systems are biologically inspired learning and optimization methods which mainly consists both the antigens and the antibodies. Artificial immune system composes characteristics of immunological learning, immunological cross-reaction and immunological memory which other evolution algorithms rarely had. The optimal solution regards as the antigen. The route candidate solutions regards as the antibodies. Antibody has three operation characteristics including affinity maturation, somatic hypermutations and immunological memory update. After many evolutions, antibodies will get optimal solutions in the limited time.

延伸閱讀