PlatformIndustriesTechnologySecurityPricingCompany

Products

Plyoq Layup & Fiber Placement Voidra Defect Sensing & NDT Curevo Cure & Consolidation Strandix Laminate & Cure Digital Twin Roboply Robotics & Handling Yieldex Yield & Orchestration Request a pilot

The autonomy loop

Seven agents and a laminate-and-cure digital twin, closing the loop from perception to bounded machine-and-line control — grounded, auditable, human-supervised.

01 PER 02 PRE 03 PLA 04 CUR 05 INS 06 CER 07 OPT CURELUM CLOSED LOOP
0
Autonomy agents
0
Digital twin
4
Autonomy levels
shadow → bounded auto
0%
Actions audited
The loop

Every stage, closed

AFP HEAD TOW WIDTH 6.35mm GAP ≤0.5mm
01

Perceive

In-process vision, laser profilometry, ultrasonic and thermographic sensing across layup, cure and NDT.

02

Predict

The digital twin and defect models predict porosity, consolidation, wrinkles and part strength before they occur.

03

Place

Autonomous AFP / ATL / robotic layup steers every tow to the qualified envelope.

04

Cure

Adaptive autoclave / OOA control holds ramp, pressure, vacuum and exotherm to target Tg.

05

Inspect

Pre-cure prediction fused with post-cure ultrasonic and thermographic NDT certifies integrity.

06

Certify

Grounded, immutable audit trails make every part traceable and airworthiness-ready.

07

Optimize

Yield, scrap, cure energy and throughput are optimized across the whole factory.

Products in the loop

Each agent owns a step

Layup & Fiber Placement

Plyoq

Autonomous fiber placement — every tow steered, every ply perfect.

Closed-loop AFP / ATL / robotic layup control: steering, tow tension, gaps, overlaps, orientation, and compaction, sensed and corrected in-process.

Explore Plyoq

Defect Sensing & NDT

Voidra

See every void — porosity, wrinkles, and FOD, sensed and certified.

Predicts porosity, wrinkles, FOD, misalignment and dry spots pre-cure, then fuses ultrasonic C-scan and thermographic evidence post-cure to certify integrity.

Explore Voidra

Cure & Consolidation

Curevo

Cure with certainty — full consolidation, target Tg, zero porosity.

Adaptive autoclave / OOA cure control — ramp, temperature, pressure, vacuum, exotherm — against physics-informed cure-kinetics and consolidation models.

Explore Curevo

Laminate & Cure Digital Twin

Strandix

Hit target strength before the cure — the composites digital twin.

GPU-accelerated simulation of placement, compaction, cure kinetics, exotherm, residual stress and strength — validates and auto-optimizes recipes before the cure.

Explore Strandix

Robotics & Handling

Roboply

Robotic hands for the composites factory — trim, drill, handle, assemble.

Vision-guided robots for ply handling, tool loading and part movement, plus robotic trimming, drilling and bonding — coordinated with the whole Curelum loop.

Explore Roboply

Yield & Orchestration

Yieldex

The factory orchestrator — maximize yield, minimize scrap and cure energy.

Factory-level optimization brain: cuOpt scheduling of AFP paths, autoclave batches, NDT routing and finishing, with one real-time view of quality and yield.

Explore Yieldex

Closed vs open loop

Why acting beats assisting

Today · open loop
  • Program a path, run it, hope
  • Inspect after cure, condemn late
  • One-size cure cycles, wasted energy
  • Defects disconnected from cause
Curelum · closed loop
  • Sense, predict, correct in-process
  • Certify integrity, trace to cause
  • Adaptive cure to target Tg
  • Every action bounded and audited
Trust before autonomy

Graduated autonomy, by design

Shadow

Curelum observes and predicts alongside your team, establishing a baseline with zero control authority.

Assist

Agents recommend layup and cure moves and predict defects and strength; engineers approve.

Bounded auto

Low-risk cells run autonomously inside qualified envelopes, escalating exceptions.

Expand

Add modules and cells across the factory as trust and data compound.

The moat

A data flywheel single tools can’t spin

Every supervised correction trains the system. More customers → more supervised outcomes → better layup, cure, porosity and strength models → more autonomy → more workflow captured → deeper integration and more proprietary data. Rare defect and new-material patterns learned at one factory transfer, IP- and ITAR-preserving, to all producers running similar parts.

  • Cross-factory porosity, wrinkle, FOD and strength corpus
  • Physics-informed placement, cure and strength surrogates
  • Federated, ITAR-isolated learning
  • Qualification-artifact and switching-cost moat
M1M2M3M4M5M6M7 FIRST-PASS YIELD % +34 PTS
TARGET ALLOWABLE PREDICTED STRENGTH (MPa) P(hit) 0.98
Digital twin

Hit strength before the cure

Strandix simulates placement, compaction, cure kinetics, exotherm, residual stress and strength — validating and auto-optimizing a recipe before any carbon or autoclave energy is committed, then pushing bounded control envelopes to Plyoq and Curevo.

Integrations

On top of the systems you run

AFP / ATL controllers

Autoclave / OOA

Ultrasonic & thermographic NDT

MES / ERP

OPC-UA / OT

Historian / PLM

SSO / RBAC

Edge fleet mgmt

Assurance

Airworthiness-ready by construction

Grounded outputs, immutable audit logs, human-in-the-loop checkpoints and bounded control envelopes are built in from day one — not bolted on. Every action is traceable to the ply, machine and cure trajectory that produced the part.

GroundingEvery output tied to real measurements
AuditImmutable per-action log
ControlBounded, qualified envelopes
Human-in-the-loopOn flight-critical decisions
IsolationITAR-ready, per-site data
StandardsAS9100 / NADCAP-ready
Outcomes

Proven on hard metrics

0pts
First-pass yield
0%
Less NDT rework
0%
Cure energy saved
<1%
Porosity

See the loop on your line

Instrument one cell, prove ROI, and expand across the factory.