DRONA Labs

The physics, graphics, and applied-AI work behind DRONA.

Everything you build on DRONA flies through a compatibility engine, a specs engine, a 3D pipeline, and an AI layer that has to stay honest about what it knows and what it doesn't. This is the research that makes those four layers worth trusting. We publish our methods because engineering tools without transparent methods don't deserve the word "engineering."

physics-engine compatibility multimodal-identification 3d-rendering PBR-materials form-factor-taxonomy evidence-chains agent-tool-use simulation autonomy-levels

Featured papers

4 papers · 2026
Engineering · open access

A Physics-Grounded Compatibility and Specs Engine for Multirotor Drone Design

We describe DRONA's deterministic, isomorphic specs engine and 12-rule compatibility system. We derive the core thrust, flight-time, noise, and TWR formulas from momentum theory with empirical stator/prop factors, and formalize the 12 universal rules as predicates over a DroneConfig type. We report agreement with published reference designs within ±15% on an 18-preset snapshot.
DRONA Labs·14 min read·Apr 2026
Applied AI

Photo-in-Chat: Grounded Multimodal Part Identification for Hardware Design

We present a two-stage identification pipeline combining a vision-language model with a catalog-grounded retrieval step over a pgvector index. We formalize the trust budget: a candidate is only committed to the build when top-1 confidence and grounded-match agree. We evaluate on 240 crowd-sourced images across seven part categories.
DRONA Labs·9 min read·Apr 2026
Methodology

A Form-Factor Taxonomy for Unmanned Systems Design Tools

A 24-entry taxonomy of unmanned aerial, surface, underwater, and hybrid systems, organized by lift principle, propulsion topology, and operator locus. We argue for form factor as a first-class type parameter in design tools, and propose a minimal-shared-ancestor discipline for compatibility rules across heterogeneous airframes.
DRONA Labs·11 min read·Coming
Computer graphics

A Procedural PBR Pipeline for Parametric Airframe Visualization in the Browser

A runtime pipeline for generating physically-based materials, normal maps, and airfoil / fuselage geometry entirely in the browser, targeting 60 fps on mid-tier mobile GPUs. We report the 19-material palette, the procedural normal-map generators (carbon-weave, brushed-metal, PCB-traces, TPU-layers), and LOD thresholds that preserve visual fidelity at near and mid-range camera distance.
DRONA Labs·12 min read·Coming

Engineering notes

shorter-form, developer-focused

Core contributors

rotating authorship

DRONA Labs is a small, rotating authorship group. We publish under the lab name rather than individual names on most papers because our work is collaborative by design and we want the research to travel with the product, not the CV.

P
Physics & simulation
Specs engine · solvers · simulation integration
G
Graphics & 3D
R3F pipeline · PBR · manifold CSG · LOD
A
Applied AI
Tool-use design · multimodal · evidence chains
D
Domain experts
Agronomy · solar · marine · SAR · firmware
How we work →
Open research

Publish with us or read our code.

Most of our methods are open. If you are doing your own research in drone physics, computer graphics for robotics, or AI-grounded hardware design, we'd love to hear from you — and cite your work where it applies.

Get in touch Follow the lab