This paper investigates optimal orbital evasion problem with considering observability performance by using a multi-objective optimization approach. The degree of observability is defined as a new performance index, which has a negative correlation with the accuracy degree of relative state estimation. A two-objective optimization model is then formulated and the NSGA-II algorithm is employed to obtain the Pareto-optimal solution set. The numerical results show that the proposed approach can effectively and efficiently demonstrate the relations among the evasive mission characteristic parameters. The proposed approach offers a novel view in solving orbital evasion problem.