To solve the global path for a biped wall-climbing robot in a 3D wall environment, a planning method combining wall transition analysis, searching of global wall sequences, and optimization of adhesion points when performing wall transitions is proposed. Firstly, to obtain the feasibility for the biped wall-climbing robot to transit between walls, the robot’s reachable workspace and the wall are simplified to transform the problem into a geometric intersection test. Then, the global wall sequence is found by using a graph search method. Finally, to obtain the optimal global path, a mathematical model is established to compute the optimal adhesion point for transiting between adjacent walls. A biped wall-climbing robot, named W-Climbot, is used for simulation. The simulation results show that the proposed method takes only 2ms on average to analyze the feasibility of transition between different walls and search the global wall sequence in a three-dimensional environment composed of 5 to 20 walls. The success rate to obtain the optimal global path is 95%, and the average time consumed is less than 4s. This method can provide optimized global paths for biped wall-climbing robots and lay a foundation for motion planning.