BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Generalisable 3D printing error detection and correction via neura
 l networks - Sebastian Pattinson\, Assistant Professor in the Department o
 f Engineering at the University of Cambridge
DTSTART:20221201T140000Z
DTEND:20221201T150000Z
UID:TALK193405@talks.cam.ac.uk
CONTACT:Fulvio Forni
DESCRIPTION:Material extrusion is the most widespread additive manufacturi
 ng method but its application in end-use products is limited by vulnerabil
 ity to errors. Humans can detect errors but cannot provide continuous moni
 toring or real-time correction. Existing automated approaches are not gene
 ralisable across different parts\, materials\, and printing systems. In th
 is talk I will discuss recent work in our lab where we train a multi-head 
 neural network using images automatically labelled by deviation from optim
 al printing parameters. The automation of data acquisition and labelling a
 llows the generation of a large and varied extrusion 3D printing dataset\,
  containing 1.2 million images from 192 different parts labelled with prin
 ting parameters. The thus trained neural network\, alongside a control loo
 p\, enables real-time detection and rapid correction of diverse errors tha
 t is effective across many different 2D and 3D geometries\, materials\, pr
 inters\, toolpaths\, and even extrusion methods.  \n\nThe seminar will be 
 held in LR5 \, Baker Building\, Department of Engineering\, and online (zo
 om): https://us06web.zoom.us/j/87986687566?pwd=MGJScmMwd2lwT0tVMHNmWmxSa05
 XZz09
LOCATION:Department of Engineering / Online (Zoom)
END:VEVENT
END:VCALENDAR