OZON Sorter Digital Twin

Web digital twin of an Ozon Tech Track 3 conveyor sorting line: continuous product playback, camera/classification stage, B/C/D routing, and author-CAD diverter motion.

Production domain: https://arhipovdan.ru

Current solution

This repository (main) contains the final web simulation:

• continuous 3D conveyor digital twin (/);
• engineering documentation page (/documentation);
• author CAD conveyor source (3d_models/conveer.FCStd);
• runtime conveyor GLB (public/models/sorter/conveyor-clean.glb);
• product STL assets (public/models/*.stl);
• camera-triggered measurement / rule-based classification (digital sensor simulation);
• B / C / D route mapping with CAD left/right diverters (−45° / +45°);
• Rapier physics: belt ≈ 1.0 m/s, fixed timestep 1/120 s, CCD on product profiles, discharge into receivers.

Main functions

1. Continuous SKU playback on a three-module CAD conveyor (clean → camera → sorter).
2. Camera-volume triggered measurement and B/C/D classification.
3. Physical diverter routing and receiver settle detection.
4. Desktop WebGL twin + mobile-capable 3D policy.
5. Canonical engineering status on /documentation.

Technology stack

Layer	Packages
UI	React, react-router-dom
3D	three, @react-three/fiber, @react-three/drei, @react-three/postprocessing
Physics	@react-three/rapier, @dimforge/rapier3d-compat
Build	Vite, TypeScript, @vitejs/plugin-react
Tests	Vitest, Playwright

See package.json for exact versions. Do not treat transitive npm packages as first-class product dependencies.

Dependencies / Node

• Node.js 20+
• npm (lockfile: package-lock.json)

Exact commands

npm ci
npm run dev          # development: http://127.0.0.1:3100
npm test -- --run    # unit tests
npm run build        # production build → dist/
npm run preview      # serve build: http://127.0.0.1:3100
npm run test:e2e     # Playwright (needs preview or PLAYWRIGHT_BASE_URL)

Active routes

Route	Purpose
/	Continuous digital-twin simulation
/documentation	Canonical product / engineering status
*	Redirects to /

Removed routes (/details, /device-test) are not part of the product.

Project structure

src/                 React app, domain logic, 3D twin, pages
public/              Runtime static assets (models, draco)
  models/            Product STL + sorter/conveyor-clean.glb
3d_models/           Author CAD (conveer.FCStd)
docs/                Engineering notes (ENGINEERING.md)
e2e/                 Playwright smoke / routes
input_info/          Official Ozon input packs (PDFs/ZIPs)
official_sources/    Classifier bounds PDF cited by code
Dockerfile           Multi-stage Vite build + nginx
docker-compose.server.yml
nginx.conf
.github/workflows/ci.yml

Environment variables

No runtime secrets are required for local demo or production static hosting.

Optional build-time identity (Docker / CI only; never commit real secrets):

Variable	Purpose
VITE_BUILD_COMMIT	Short git commit in /version.json
VITE_BUILD_BRANCH	Branch name in /version.json
VITE_BUILD_RELEASE	Release label in /version.json

Do not commit .env files. Local npm run build derives identity from git when available.

Active assets

Path	Role
3d_models/conveer.FCStd	Author CAD (canonical source)
public/models/sorter/conveyor-clean.glb	Active runtime conveyor
public/models/*.stl	Product models
input_info/*	Official Ozon packs (PDFs/ZIPs)
official_sources/doc-1783095831.pdf	Official classifier bounds source cited by code

Frozen checksums (SHA-256)

3d_models/conveer.FCStd
90c1844a4ca05e26def783d6130fc4b993430dde14307534ef8fbb21c9fac2e6

public/models/sorter/conveyor-clean.glb
1dc7a8d7891bfe756e277ad5368df74cb73410156b2fe0f92845afb8a56f285a

Classifier vs real CV

Web twin (main): classification is a digital sensor simulation. When a product enters the camera scan volume, dimensions / roundness are measured in the twin and routed with the same B/C/D rules as Track 3 (src/domain/classifier.ts, src/domain/cameraClassification.ts). This is not a neural network and is not wired to a live RealSense camera in production.

Real CV (parallel prototype, not merged into the web app): preserved on branch drho1y-mvp_1 under vision_classifier/ (OpenCV + RealSense D415 depth → L×W×H + circle_ratio → B/C/D → MQTT). No PyTorch/YOLO weights are stored in git.

git fetch origin
git switch drho1y-mvp_1
cd vision_classifier
# see vision_classifier/README.md and START.md
./demo.sh                  # browser HUD demo
.venv/bin/python test_classify.py
./run.sh --preview         # full pipeline on hardware (Orange PI + RealSense)

Integration status: WORKING_PROTOTYPE on drho1y-mvp_1, not integrated into https://arhipovdan.ru.

Sorting logic

CURRENT_IMPLEMENTATION_VERIFIED_IN_CODE (src/domain/classifier.ts, src/domain/pusherMotion.ts, unit tests).

Category	Physical route	Active CAD diverter
B	STRAIGHT	none
C	PHYSICAL_LEFT	LEFT, −45°
D	PHYSICAL_RIGHT	RIGHT, +45°

Frozen diverter timing / planes (verified in code + tests):

• rotation duration: 0.50 s
• opening safety margin: 0.15 s
• contact plane S: 1.0538 (fallback hinge geometry ≈ 1.0536)
• clear plane S: 1.6000

Classifier bounds (code + official_sources/doc-1783095831.pdf reference; PDF not re-parsed in this doc pass):

• dimensions strictly > 10×10×10 mm and < 450×320×320 mm
• circular when K > 0.8
• check order: dimensions → C, else circular → D, else B

Physical simulation

• Fixed timestep 1/120 s, max 4 substeps, gravity [0, −9.81, 0]
• Belt target speed 1.0 m/s (supported-body velocity coupling)
• CCD enabled on active product physics profiles
• Single dynamic product rigid body through spawn → junction → receiver settle
• LEFT/RIGHT CAD diverters: kinematicPositionBased colliders synced to CAD yaw
• Discharge edge ends belt support; gravity fall into receivers
• Receiver volumes are sensors for completion

Details: docs/ENGINEERING.md, /documentation.

Validation

npm ci
npm test -- --run          # currently 245 unit tests
npm run build
# focused E2E (with preview on :3101):
npx vite preview --host 127.0.0.1 --port 3101 &
PLAYWRIGHT_BASE_URL=http://127.0.0.1:3101 npm run test:e2e -- e2e/routes.spec.ts e2e/smoke.spec.ts

• Two-page routing: / + /documentation
• CAD / GLB checksums: verified against values above
• Diverter frozen angles / duration: covered by unit tests
• Junction / discharge / camera classification: covered by domain tests (engineering-derived, not production-calibrated)

Production deploy (this host)

Nginx terminates TLS for arhipovdan.ru and proxies to Docker owl-web-1 on 127.0.0.1:3100 (docker-compose.server.yml + Dockerfile).

cd /opt/arhipovdan/app   # or your checkout
COMMIT=$(git rev-parse --short HEAD)
BRANCH=$(git rev-parse --abbrev-ref HEAD)
RELEASE=$(date -u +%Y%m%d-%H%M%S)
docker compose -p owl -f docker-compose.server.yml build \
  --build-arg "BUILD_COMMIT=${COMMIT}" \
  --build-arg "BUILD_BRANCH=${BRANCH}" \
  --build-arg "BUILD_RELEASE=${RELEASE}"
docker compose -p owl -f docker-compose.server.yml up -d --force-recreate
curl -s https://arhipovdan.ru/version.json

Repository vs cloud materials

Material	In git	Cloud expected
Source, README, docs, configs, tests	yes	—
Runtime GLB/STL used by the deployed app	yes (required for deploy)	recommended mirror
Author CAD conveer.FCStd (~5 MB)	yes	recommended upload
Official PDFs/ZIPs under input_info/, official_sources/	yes	optional mirror
Presentation / video demonstration	not in repo	required for platform field
ML weights	none (CV is classical OpenCV depth)	N/A

Confirmed public presentation / video / cloud-folder URLs are not stored in this repository. Add them in the platform submission field when available.

Презентация / видеодемонстрация: links must be supplied by the owner (not invented here).

Current limitations

• Web classifier is a digital sensor simulation, not live RealSense inference.
• Product profiles are engineering-derived, not production-calibrated.
• Author CAD horn / complete transmission is absent or incomplete in the active GLB (AUTHOR_CAD_INCOMPLETE).
• Official compliance claims are limited by the missing extracted task PDF and by not re-parsing PDFs in every doc pass.
• Generated screenshots, videos, Gate stage folders, and tool out/ trees are not canonical.

Repository policy

• Keep author CAD, active runtime assets, official sources, build configs, and tests that protect active behavior.
• Generated outputs and historical media are not product truth.
• Real CV remains on drho1y-mvp_1; do not force-merge it into main without a validated web integration.
• Rollback point for the pre-cleanup web baseline: git 4413f01 on dan_branch history.

See also: /documentation in the running app, and docs/ENGINEERING.md.