Rigel

01The problem

Aerospace design workflows suffer from severe friction between high-level engineering intent and low-level computational geometry synthesis.

Engineers waste days manually modeling CAD brackets, configuring meshes, and guessing parameters, while generative AI hallucinations cannot be trusted in safety-critical airframes.

02Why it has to exist

Every aerospace structural component must satisfy rigorous bending, shear, and stress-concentration limits under single or multi-axial load cases (e.g. Al6061-T6 under 4 kN).

'AI handles ambiguity. Math handles precision. These roles never swap.'

03What we learned first

Grounded in foundational research on Parametric Flight Surface Synthesis and The Derivation Engine framework.

Inquires how first-principles physical equations (Shigley Eq 6-2) translate directly into B-Rep geometry and verifiable FEA convergence.

04How it works

The Derivation Engine: Local LLaMA 3.2 3B parser ingests plain-English load specs; Shigley equations derive exact dimensions; PicoGK synthesizes geometry; Gmsh generates Tet10 volume meshes; CalculiX static FEA validates Von Mises stress and factor of safety (FOS); SHA-256 cryptographic chain seals audit trail.

05How it's built

Rust control plane with C/C++ FFI bindings (Gmsh C API, CalculiX subprocess).

Outputs four mandatory production artifacts every run: bracket.step (production geometry), bracket.dxf (CNC ready), report.pdf (stress report with red disclaimer), and audit.json (SHA-256 tamper-evident hash chain).

• Rust
• Shigley Physics
• PicoGK Kernel
• Gmsh (Tet10)
• CalculiX FEA
• SHA-256 Audit Chain
• Al6061-T6 Sizing

06What we've tested

Evaluated on laboratory testbed exp-01 and physical load-to-failure testing on a workshop press at PSG College of Engineering, Coimbatore.

Target: physical failure loads agree with FEA predictions within 15%. All deviations recorded publicly in FAILURES.md. On a standard 8-core workstation, the full derivation pipeline executes in 8-12 minutes without CAD hallucination. Iterative 2D slice decomposition reduced constraint latency by ~80% prior to full volume meshing.

07Where it stands

Stage 1 MVP in active development.

Six-file architecture discipline strictly enforced; core solver lattice running local-only computation with verified Shigley equations.

08What's next

Integrating non-linear gradient descent escape algorithms to avoid local aerodynamic minima during sudden airfoil taper shifts and multi-load fastener patterns.