BEGIN:VCALENDAR
VERSION:2.0
PRODID:Linklings LLC
BEGIN:VTIMEZONE
TZID:Asia/Tokyo
X-LIC-LOCATION:Asia/Tokyo
BEGIN:STANDARD
TZOFFSETFROM:+0900
TZOFFSETTO:+0900
TZNAME:JST
DTSTART:18871231T000000
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20260817T171530Z
LOCATION:Hall B5 (1)\, B Block\, Level 5
DTSTART;TZID=Asia/Tokyo:20241205T105900
DTEND;TZID=Asia/Tokyo:20241205T111300
UID:siggraphasia_SIGGRAPH Asia 2024_sess127_papers_533@linklings.com
SUMMARY:Learning Based Toolpath Planner on Diverse Graphs for 3D Printing
DESCRIPTION:Yuming Huang, Yuhu Guo, and Renbo Su (University of Manchester
 ); Xingjian Han (Boston University); Junhao Ding (Chinese University of Ho
 ng Kong); Tianyu Zhang, Tao Liu, and Weiming Wang (University of Mancheste
 r); Guoxin Fang and Xu Song (Chinese University of Hong Kong); Emily Whiti
 ng (Boston University); and Charlie Wang (University of Manchester)\n\nThi
 s paper presents a learning based planner for computing optimized 3D print
 ing toolpaths on prescribed graphs, the challenges of which include the va
 rying graph structures on different models and the large scale of nodes & 
 edges on a graph. We adopt an on-the-fly strategy to tackle these challeng
 es, formulating the planner as a Deep Q-Network (DQN) based optimizer to d
 ecide the next `best' node to visit. We construct the state spaces by the 
 Local Search Graph (LSG) centered at different nodes on a graph, which is 
 encoded by a carefully designed algorithm so that LSGs in similar configur
 ations can be identified to re-use the earlier learned DQN priors for acce
 lerating the computation of toolpath planning. Our method can cover differ
 ent 3D printing applications by defining their corresponding reward functi
 ons. Toolpath planning problems in wire-frame printing, continuous fiber p
 rinting, and metallic printing are selected to demonstrate its generality.
  The performance of our planner has been verified by testing the resultant
  toolpaths in physical experiments. By using our planner, wire-frame model
 s with up to 4.2k struts can be successfully printed, up to 93.3% of sharp
  turns on continuous fiber toolpaths can be avoided, and the thermal disto
 rtion in metallic printing can be reduced by 24.9%.\n\nRegistration Catego
 ry: Full Access, Full Access Supporter\n\nLanguage Format: English Languag
 e\n\nSession Chair: Lin Lu (Shandong University)\n\n
URL:https://asia.siggraph.org/2024/program/?id=papers_533&sess=sess127
END:VEVENT
END:VCALENDAR