Due to the need for driving routes that emphasize scenery, it is an important research topic to explore various routes while passing through predetermined spots. In this study, we propose a method that expresses the spots that should be passed through by a virus and generates a route that satisfies the constraints on passing through the spots by infection. In addition, by performing tournament selection and elite selection in each cluster obtained through clustering, generational change while maintaining diverse routes is realized. Through evaluation experiments targeting two urban networks in Japan, we confirm the effectiveness of the proposed method in terms of route diversity and constraint sufficiency by comparing it with a regular genetic algorithm without virus infection and clustering.