Abstract: The automatic robots are gradually applied in the process of unmanned coal mining,hazard detection and disaster relief. According to the characteristics of small confined space,complex obstacle distribution and uneven floor in underground coal mine,a three-dimensional (3D) path planning method for robot in underground local complex space is proposed based on hybrid ant colony and artificial bee colony algorithm. The proposed method can generate the initial paths simply and extend the search range of new feasible path,which effectively solves the problem that ant colony algorithm may trap into the local optimal solution and the iteration times of the artificial bee colony algorithm may be numerous. The B-spline interpolation is applied to the 3D broken path line to generate the optimal continuous smooth path,which is beneficial for robot to travel smoothly. Simulation results show that the robot can plan its path ef- fectively in underground local complex space with the proposed path planning method based on hybrid ant colony and artificial bee colony algorithm.