Print-Scan-Adapt: Feedback-Driven Bead Geometry Control in Pellet Extrusion

Registration Link

Workshop Registration opens September 8th, 2025! Click the link above to be notified when registration becomes available.

Location : FIU

Workshop Team

Aldo Sollazzo, LAMÁQUINA
Hritik Thumar, LAMÁQUINA

Workshop Description

Print–Scan–Adapt is a research-driven workflow developed at LAMÁQUINA that integrates sensing, data analysis, and adaptive control into large-scale 3D printing. Traditional extrusion processes often produce inconsistent results, with bead width and layer accuracy shifting due to speed, acceleration, and material flow. By combining real-time scanning with robotic parameter mapping, this workflow transforms fabrication challenges into data-driven design opportunities.

This workshop brings the research to life in a hands-on format, exploring feedback-driven control in large-scale pellet-based 3D printing. Pellet extrusion offers a sustainable alternative to filament deposition, supporting recycled polymers and industrial-scale output, yet the process is inherently unstable. Participants use a flange-mounted Intel RealSense depth camera to measure bead width during printing, correlating this data with robot speed and extrusion parameters via KUKA|prc. The resulting mapping model predicts and adapts deposition outcomes, allowing participants to actively control material behavior.

Print–Scan–Adapt emphasizes computation as a tool for resilience at the material-process interface, turning unpredictable extrusion into a design opportunity. By deliberately guiding over- and under-extrusion, participants create patterned modules where material variability becomes an intentional aesthetic. The final outcome is a collective prototype — such as a column or arch — assembled from modules whose surface textures directly reflect the process feedback, demonstrating how sensing, computation, and design converge in large-scale additive fabrication.

Key Learning Outcomes

Understand Pellet-Based Additive Manufacturing: Identify and analyze the challenges of pellet extrusion, including flow instability, bead width variation, and die swell.
Integrate Sensing for Real-Time Feedback: Set up and operate a scan-to-feedback workflow using Intel RealSense and Grasshopper, correlating robotic parameters with measured bead geometry.
Develop Predictive Toolpath Models: Build width-mapping models that translate target bead sizes into adjustable toolpath parameters and control variable deposition as a design element.
Fabricate Adaptive, Data-Driven Modules: Apply image-sampler logic and adaptive control to produce modules demonstrating material responsiveness and surface variation.
Collaborate on Collective Assemblies: Contribute to a final prototype that embodies resilience, material intelligence, and computational design.

Workshop Schedule

Day 1 – Calibration & Data Collection

Time: 10:00 am – 5:00 pm (w/ 1-hour break)

Welcome + overview of ACADIA theme & workshop goals
Introduction to pellet extrusion and feedback workflows
Printing test coupons: straight lines, arcs, corners at varying speeds
Scanning workflow: bead width extraction with RealSense
Building initial speed vs width dataset

Deliverables:

Set of calibration prints
Logged dataset correlating speed with bead width
Preliminary visualization of width vs. speed graph

Day 2 – Mapping & Design Integration

Time: 10:00 am – 5:00 pm (w/ 1-hour break)

Developing a mapping function: target width → robot parameters
Introducing image-sampler logic in Grasshopper
Designing variable-width toolpaths based on images/patterns
Pilot printing of patterned strips with scan validation

Deliverables:

Fitted width-mapping function
First patterned test strips with visible variation in bead width
Scan data comparing intended vs. achieved bead geometry

Day 3 – Prototyping & Assembly

Time: 10:00 am – 5:00 pm (w/ 1-hour break)

Generating toolpaths for final modules (column/arch components)
Printing with variable-width toolpaths; scanning to validate
Assembly of modules into collective prototype
Wrap-up: discussion of resilient workflows, sustainability, and design expression

Deliverables:

Individual modules printed with intentional bead width variation
Scans validating controlled over-/under-extrusion
Collective prototype (assembled column/arch)
Group reflections on outcomes and future applications

Total Workshop Time: ~21 hours over 3 days

Participant Information

Audience: Students, researchers, and professionals in architecture, design, and digital fabrication
Skill Level: Intermediate. Prior knowledge of Rhino/Grasshopper is helpful but not required
Group Size: 15–20 participants