ORCA: Open RF Integrated Circuit Automation¶
ORCA (Open RF Integrated Circuit Automation) builds surrogate models of RF integrated circuit components by combining parametric GDS layout generation, full-wave electromagnetic simulation, and neural network training and export in one automated pipeline.
Authors
Gianluca Simone, David Lurz, Martin Grund, Fabian Schneider, Michael Loose, Sascha Breun, Manuel Koch, Robert Weigel, Norman Franchi
Published
SBCCI 2026 (conference contribution)
Main Capabilities¶
- Parametric GDS layout generation via gdsfactory.
- Full-wave electromagnetic simulation via Palace.
- Automatic dataset construction from simulation results.
- Neural network training (PyTorch) with feature engineering pipelines.
- ONNX export for portable, simulator-ready surrogate models.
- GUI workflow and script-first workflow.
- Pre-trained models and geometry examples included.
Start Here¶
- New users: go to Installation.
- First successful run: Quickstart.
- Understand the stages: Pipeline Stages.
- Bring your own geometry: Custom Classes.
- Publish a trained model: Sharing Models on Hugging Face.
- API details: see the source and docstrings in
src/orca/.
Core Concepts¶
ORCA automates the full loop from geometry to a trained, exported surrogate model.
Pipeline Workflow¶
flowchart LR
A[Geometry Class] --> B[GDSGenerator]
B --> X[DRCChecker]
X --> C[GDSConverter]
C --> D[PalaceSimulator]
D --> E[S-Parameter Dataset]
E --> F[ModelTrainer]
F --> G[OnnxExporter]
G --> H[ModelTester]
H --> I[.onnx Surrogate]
Pipeline Stages¶
| Stage | Purpose |
|---|---|
GDSGenerator |
Samples geometry parameters and writes GDS layout files |
DRCChecker |
Snaps layouts to the manufacturing grid and drops those violating the SG13G2 design rules or with port markers off their metal |
GDSConverter |
Converts GDS files to Palace-compatible mesh inputs |
PalaceSimulator |
Runs full-wave EM simulations and stores Touchstone results |
ModelTrainer |
Trains a PyTorch neural network on the simulation dataset |
OnnxExporter |
Exports the trained model to portable ONNX format |
ModelTester |
Evaluates prediction accuracy against held-out simulation data |
Geometry Class¶
A geometry class defines everything ORCA needs to sample and simulate a component:
- Parameter ranges via
InputParameterIteratorofRangeParameter/ChoiceParameterobjects (min/max, step, data type and uniform or log sampling per parameter; space-filling Sobol' (default), Latin hypercube, random or grid picking, with samples on the boundary of the box). - GDS generation via
create_gds_file(name, output_path, params). - Basis expansions of the model inputs via
BasisExpansion(e.g. a Chebyshev expansion of frequency), chosen onModelTrainerand shared by every architecture. - Dataset type (e.g.
GeoToSParamDatasetSingleFrequency). - Model architecture via
OrcaModelsubclasses registered withregister_model, chosen onModelTrainer(model=...). - Hyperparameter search space via the model's
hyperparameter_search_space()— Optuna distributions per tunable parameter, merged with the trainer's and the basis expansion's.
Inputs and Outputs¶
- A geometry class (`BaseGeometry` subclass) with parameter definitions.
- A stackup XML file describing the physical layer stack.
- A Palace simulation configuration file (`.simcfg`).
- A Palace executable (local or containerised via Apptainer).
- Touchstone (`.sNp`) S-parameter files per simulated geometry variant.
- Trained PyTorch model checkpoint.
- Exported ONNX surrogate model (`.onnx`).
- Full run context stored under `output/<geometry_name>/`.
Related Projects¶
- COBRA consumes ORCA surrogate models for circuit-level RFIC optimization.
- Palace is used as the full-wave EM solver backend.
- gdsfactory generates the parametric GDS layouts.
- gds2palace converts GDS layouts into Palace simulation setups.
If you use ORCA in research, see citation information in the repository README.