Path planning of dual-arm robot is very complicated, and the efficiency of traditional algorithms is low. To solve this problem, the FT-heaps-A* algorithm is presented in this letter. A new parameter ‘FT’ is introduced to avoid the invalid path nodes in the high-dimensional environments. Then, the time required to maintain the open table structure is reduced by splitting the open table. Thus, the efficiency of path planning for the single manipulator is improved. Based on the above improved A* algorithm, a time series A* (TSA*) algorithm is proposed to plan the obstacle avoidance path of the dual-arm robot. The simulation results show that the proposed algorithm has higher path quality and efficiency than the traditional algorithms.