Abstract: The styling feature line, a typical element of automobile styling gene，is selected as research object. Based on genetic algorithm, the thought and method flow of automobile styling gene evolution driven by users’ expectation image is proposed. The adjectives of users’ expectation image are obtained through investigation and benchmark automobile analysis. The key image words are obtained through cluster analysis and characterized quantitatively. Automobile side profile line is taken as research object and described quantitatively by using Bezier curves. The case samples of initial population that satisfy the users’ expectation image are obtained through several methods and they are translated to genotype. After constructing the mechanism of artificial fitness evaluation based on the users’ expectation image adjectives, the progeny populations according with users’ expectation image are obtained through crossover and mutation operations. Thus the target of automobile styling gene evolution driven by users’ expectation image is attained. A case of two-box mini auto styling evolution design is given, which demonstrates that the method is rational and feasible..

Key words: Automobile styling, Evolution, Expectation image adjective, Genetic algorithm, Styling gene