Abstract: In view of getting alternatives of the car styling design, the method based on modeling parts and Adaptive Interactive Genetic Algorithm（AIGA） is introduced. In the AIGA, self adaptive probability of crossover is based on cloud models, and mutate probability is varied depending on similarity of individuals and evolutionary generation. An AIGA-based aided system prototype for the car styling design is constructed and realized. The simulation experiment indicates that the proposed method can obtain contented alternatives of automobile form, reduce user fatigue, and have higher performance.

Key words: car styling, interactive genetic algorithm, cloud models, user fitigue

摘要： 针对汽车造型设计的方案求解问题，提出基于造型件和利用自适应交互式遗传算法的汽车造型设计方法。在自适应交互式遗传算法中，利用云模型产生自适应交叉概率，而利用个体之间的相似性和进化代数产生自适应变异概率。通过汽车造型设计实验，表明所提出求解算法可以获得满意的汽车造型设计方案，能够降低用户疲劳，具有较高的运行性能。

关键词: 汽车造型, 交互式遗传算法, 云模型, 用户疲劳