Aiming at the parameter design problem of non-singular terminal sliding mode (NTSM) terminal guidance law with falling angle constraint, a parameter optimization method based on genetic algorithm is proposed. Firstly, a two-dimensional guidance model is established. The NTSM terminal guidance law with falling angle constraint is derived by selecting the nonlinear sliding mode surface and the double power reaching law. On this basis, the guidance parameters to be designed are connected with the guidance effect, and the guidance parameter optimization process is transformed into a process of searching for optimal solutions based on genetic algorithm. In response to the needs of flight performance and reachable range during terminal guidance with falling angle constraint, the acceleration command value and the number of reachable falling angles are selected as evaluation functions respectively, and a parameter optimization method is designed to minimize the acceleration command and maximize the falling angle range. This method can obtain the optimal parameters that adapt to the optimization goal and avoid the inefficiency and limitations of manual design. Finally, comparative simulation results are presented to verify the feasibility, effectiveness and superiority of the proposed method.