A global path planning method, based on the improved artificial fish swarm algorithm(IAFSA) and MAKLINK graph,is proposed to solve the global path planning problem in the two-dimensional static environment．Lorentzian function and normal distribution function are chosen as adaptive operators of step and visual, the exponential decreasing inertia weighting factor is also introduced, which can improve the convergence speed and accuracy of the AFSA algorithm. The MS algorithm is combined with the IAFSA algorithm to calculate for two steps. The optimal path optimized by the IAFSA algorithm is selected as the global optimal path, which solves the problem of that the previous algorithm can only get the approximate global optimal path in the MAKLINK graph. The simulation results show the feasibility and effectiveness of the proposed improved algorithm.