A Reinforcement Learning-based Path Planning Method for Complex Thin-walled Structures in 3D Printing

Path planning is an important part of the 3D printing process. The optimized path planning method can improve not only effect of the molding but also the efficiency of printing process. However, traditional path planning methods are not satisfactory in 3D printing, especially when printing the entities with complex thin-wall structures. We propose an intelligent path planning method named Q-Path, based on reinforcement learning for complex thin-walled structures. We first convert the path planning task to a full-path traversing problem. Then we use the Q-learning algorithm to find the optimal solution with the constraints of 3D printing, such as the minimum number of lifts and turns of the print head. Experimental results show that the proposed methods are superior to the traditional methods in printing complex thin-walled structures.