Learning Based Toolpath Planner on Diverse Graphs for 3D Printing

• DOI: 10.1145/3687933
• Publication Date: 2024-11-19T15:46:04Z
• Authors: Yuming Huang; Yuhu Guo; Renbo Su; Xingjian Han; Junhao Ding; Tianyu Zhang; Tao Liu; Weiming Wang; Guoxin Fang; Xu Song; Emily Whiting; Charlie Wang
• Keywords: Planner; Stepping stone

This paper presents a learning based planner for computing optimized 3D printing toolpaths on prescribed graphs, the challenges of which include varying graph structures different models and large scale nodes & edges graph. We adopt an on-the-fly strategy to tackle these challenges, formulating as Deep Q-Network (DQN) optimizer decide next 'best' node visit. construct state spaces by Local Search Graph (LSG) centered at graph, is encoded carefully designed algorithm so that LSGs in similar configurations can be identified re-use earlier learned DQN priors accelerating computation toolpath planning. Our method cover applications defining their corresponding reward functions. Toolpath planning problems wire-frame printing, continuous fiber metallic are selected demonstrate its generality. The performance our has been verified testing resultant physical experiments. By using planner, with up 4.2k struts successfully printed, 93.3% sharp turns avoided, thermal distortion reduced 24.9%.