基于GEP和L-系统的植物形态模拟  

PLANT MORPHOLOGY SIMULATION BASED ON GENE EXPRESSION PROGRAMMING AND L - SYSTEM

作　　者：代晓琳[1] 马学强[1,2] 王甜甜[1] 

机构地区：[1]山东师范大学信息科学与工程学院,济南250014 [2]山东省分布式计算机软件新技术重点实验室,济南250014

出　　处：《山东师范大学学报（自然科学版）》2016年第1期37-42,共6页Journal of Shandong Normal University(Natural Science)

基　　金：山东省高等学校科技计划项目（J14LN09）.

摘　　要：传统的 L -系统规则都是通过先验知识人工提取的出来的，并且通过 L -系统文法结合海龟解释实现植物的可视化模拟，但是传统 L -系统规则的提取是非常耗时，尤其是在大型场景建模时，这种方法的效率非常低。综合考虑这些问题，本文提出了基于基因表达式编程和 L -系统结合的植物三维形态建模方法，这种方法结合了遗传算法并行快速搜索最优解的特点和开放式L 系统通过重写规则建模的优势，并且不需要人工提取规则参数，而是通过遗传操作自动生成规则，然后根据设计的适应度函数决定种群个体是否传到下一代，最后生成我们需要的目标植物，实验证明提出的模型可以快速的实现植物的可视化。The rules of traditional L - system are extracted by the prior knowledge manually. The plant simulation is visualized by combing the turtle interpretation and iterative grammar system. But the extraction of rules is very time - consuming. Its efficiency is especially low in large - scale scene modeling. Considering these problems,we put forward the modeling method of the 3D plants combining genetic expression programming and L - system,which actually combines the characteristics of the genetic algorithm that can parallelly and quickly search for optimal solution,and the advantage of open L - system that can model by rewriting rules. And the rule parameters do not need to be extracted artificially. Instead,these rules are generated by the genetic operation automatically,and then whether individual species passes to the next generation is determined according to the designed fitness function. Finally,the target plants are generated . Experiments show that the proposed method is more effective in realizing the visualization of plants.

关 键 词：L 系统 基因表达式编程 适应度函数 虚拟植物 遗传算法 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]