For simulating crawler crane lifting, a kinematic model is established by an improved DH (Denavit-Hartenberg) method. The regular geometry enveloping crane boom, the lifted object and the obstacles are modeled in the lifting task. In order to improve the lifting efficiency and safety, a crane lifting path planning algorithm is proposed, based on the bi-directional RRT (Rapidly Random Exploring Tree) algorithm where the principle of artificial potential field method is introduced. At the same time, through applying the B spline curve, the strategy of deleting redundant nodes of the path is generated to reduce the frequent action switching of the crane. Then the sequence of crawler crane lifting actions is solved by inverse kinematics. The simulation experimental results for the traditional RRT algorithm, the bi-directional RRT algorithm and the improved algorithm show that the improved algorithm proposed in this paper can effectively reduce the path search time and path length, and achieve the smooth crane lifting action sequence.