The global path-planning method for MR is studied based on the characteristics of ant colony algorithm. The environment information of a plane on which mobile robot (MR) works is expressed by the grid method. The main parameters used by ant colony algorithm, such as the number of ant m, arouse factor α, expected arouse factor βand information element hangover coefficients ρ, which affect the result of path planned, including the length of the path planned and the efficiency planning path, are simulated. It is found from the simulation result that the better global path can be gotten. The length of the path planned is shorter and the efficiency of planning path is higher when the parameters m,α,βand ρ are matched better. Based on the results of above simulation, the path is planned for MR, which works on two different work-plane using ant colony algorithm with the best matching parameters obtained from the simulation. The accuracy of theoretical research is verified by the global path planned for MR working under the different work-plane. 