Abstract
We present an Evolutionary Planning and Replanning Algorithm, capable of producing Bounded Suboptimal solutions in an Anytime fashion. Combining Genetic Algorithm with anytime approach is uncommon. Anytime Genetic algorithm combines the benefits of an Anytime and an Evolutionary Algorithm to efficiently provide solutions to complex, Dynamic Search Problems. The results appear promising for shorter horizon problems, while for large horizon the search tends to behave like a standard GA
Publisher
Trans Tech Publications, Ltd.
Reference8 articles.
1. Zilberstein, Shlomo, Using Anytime Algorithms in Intelligent Systems. (Online) Available: http: /anytime. cs. umass. edu/shlomo/papers/aimag96. pdf.
2. Grass, Joshua. "Reasoning about Computational Resource Allocation[online]. Available: http: /www. acm. org/crossroads/xrds3-1/racra. html.
3. S. Zilberstein and S. Russel, Approximate reasoning using anytime algorithms, in Imprecise and Approximate Computation, S. Natarajan, Ed. Norwell, MA: Kluwer, 1995, p.43–62.
4. Chu-Hsing Lin; Chen-Yu Lee; Jung-Chun Liu; Hao-Tian Zuo; , Investigations of Factors Affecting the Genetic Algorithm for Shortest Driving Time, " Soft Computing and Pattern Recognition, 2009. SOCPAR , 09. International.
5. L. Wu,H. Wang, F. Lu, etc. An Anytime Algorithm based on Modified GA for Dynamic Weapon-Target Allocation Problem. Accepted by WCCI2008, to be held at HongKong, China, June, (2008).