With the increasing level of urbanization and electricity consumption in China, the problem of automated power grid route planning has gradually become a research focus for researchers. This paper first processes the maps obtained through aerial photography into vector graphics based on a deep neural network. Then, on the basis of the vector graphics, a genetic sequence of the multi-objective evolutionary algorithm based on decomposition (MOEA/D) is constructed. The algorithm takes the total length of the construction route and the estimated construction cost as the two objectives to be optimized. At the same time, a fitness evaluation function is constructed based on the weight relationship between the objectives and the segmentation method of the problem. Finally, a series of Pareto optimal solutions for the power grid laying routes are obtained. The experimental results of this paper show that even when the channel road surface is relatively complex, the algorithm can still achieve good results and has strong robustness.