Closed-loop AFP / ATL control
Real-time steering, tension, gap/overlap and compaction control across placement heads.
Layup & Fiber Placement
Plyoq
Autonomous fiber placement — every tow steered, every ply perfect.
A factory-edge agent that runs AFP, ATL and robotic layup in a closed perceive–decide–act loop — steering every tow, closing every gap, laying a defect-free laminate.
0mm
Tow width sensed
gaps & overlaps ≤ 0.5mm
0%
In-process coverage
every course, every ply
0×
Faster steering correction
vs manual head tuning
0
Blind courses
closed-loop, not open-loop
The problem
Why this step condemns parts
AFP and ATL are the throughput engines of modern composites, yet they still run open-loop: a technician programs a tool path, runs it, and hopes. Steering wrinkles, tow gaps and overlaps, misaligned plies and FOD accumulate silently across a stack already carrying tens of thousands of dollars of prepreg — while rate demand climbs and defect-free placement depends on scarce, cleared technicians tuning heads by feel.
Curelum closes the loop
Autonomous fiber placement — every tow steered, every ply perfect.— Plyoq, part of the Curelum autonomy platform
How it works
A closed perceive → decide → act loop
No open-loop hoping. Plyoq senses, reasons against the qualified envelope, acts within bounds, and logs every action.
Sense the course
Vision and laser profilometry measure tow width, gaps, overlaps, steering wrinkles, compaction and fiber angle as each course is laid.
Reason against the recipe
The agent compares live geometry to the Strandix-optimized layup envelope and predicts defect risk before it compounds.
Correct in the loop
Steering, tow tension, heat and compaction are adjusted continuously — with graduated autonomy and human-in-the-loop checkpoints on flight-critical courses.
Log every action
Every machine action is written to an immutable, assurance-grade audit trail tied to the exact ply and part.
Capabilities
What Plyoq does on the line
Fiber-angle & orientation lock
Holds ply orientation and steering radii to the qualified envelope, course by course.
In-process defect prevention
Predicts wrinkles, gaps, overlaps and FOD before they are laid — not after cure.
Graduated autonomy
Shadow → assist → bounded auto-control, escalating exceptions to the technician.
Multi-material ready
Prepreg, dry-fiber and thermoplastic tow behaviors learned and generalized.
Airworthiness audit trail
Immutable per-course action log for AS9100 / NADCAP traceability.
Read the process
Instrumented, not guessed
Every Plyoq readout maps to a real measurement on the floor — tow geometry, cure trajectory, C-scan integrity, or predicted strength. The interface is precision test equipment: numbers you can certify, not a dashboard of vanity charts.
- Grounded, auditable outputs on every action
- Bounded control envelopes from the Strandix twin
- Human-in-the-loop on flight-critical decisions
- Immutable audit trail for AS9100 / NADCAP
Specifications
Built for the composites floor
Plyoq deploys on the factory edge and integrates with the systems you already run, under your security and export controls.
| Machines | AFP, ATL, robotic and gantry layup cells |
|---|---|
| Sensing | Machine vision, laser profilometry, thermal |
| Control latency | Bounded factory-edge inference on Jetson / Holoscan |
| Materials | Carbon prepreg, dry-fiber, thermoplastic tow |
| Autonomy | Shadow / assist / graduated auto with HITL |
| Integrations | AFP OEM controllers, MES, OPC-UA, Strandix twin |
Integrations
Fits your stack, under your controls
AFP / ATL controllers
Autoclave / OOA
Ultrasonic & thermographic NDT
MES / ERP
OPC-UA / OT
Historian & PLM
Assurance first
Trust before autonomy
Plyoq runs in shadow, then assist, then bounded autonomy — grounded, auditable, and human-supervised at every airworthiness-critical checkpoint.
The platform
Stronger together across the loop
FAQ
Questions engineers ask
Does Plyoq replace our AFP machine?
No. Plyoq is a control and perception layer that runs on top of your existing AFP / ATL cells through the machine controller and OEM interfaces, adding closed-loop sensing and bounded correction.
How does it avoid unsafe machine actions?
Every action is bounded by qualified control envelopes from Strandix, gated by graduated-autonomy levels and human-in-the-loop checkpoints, and written to an immutable audit trail.
What does it need to start?
A wedge cell instrumented in shadow mode to establish a baseline, plus controller and MES access under your security / ITAR controls.
Put Plyoq on your line
Start with one cell, a hard success metric, and a paid pilot. Prove ROI, then expand across the loop.