Home/Open Crowdfunding/AI Smart Aquarium
🐠

AI Smart Aquarium · 90 days without losing a fish, even for beginners

All-in-one AI aquarium: AI-vision fish counting, dead-fish detection and activity scoring; an LLM care assistant with medication-dosage guardrails; an independent safety MCU hard-safety chain. Basic edition (ESP32-S3) ¥299–499, AI edition (RK3588S) ¥999–1699 — open for crowdfunding.

  • 🎯Crowdfunding goal from ¥20,000
  • 💰Basic ¥299–499 · AI edition ¥999–1699
Current progress · updated 2026-0920%
🔧 Design frozen · Algorithm smoke-passed · Pre-MVP
CrowdfundingOpen · goal from ¥20,000
Status● Live — accepting backers
📞 19040667970 Back this project
Highlights

What it brings you

👁️

AI vision inventory

Fish counting and in-tank inventory — an instant alert if one goes missing; dead / dying fish detection at first light.

📊

Activity index

A 7-day self-learning behavior baseline; abnormal activity warns before water quality collapses.

🤖

LLM care assistant

Aquarium knowledge RAG linked to your tank's live data, with medication-dosage guardrails; runs offline.

🛡️

Independent safety MCU

A dedicated safety chain: even a main-SoC crash cannot cause overheating; water changes never exceed the 'confirm' permission.

🎬

Highlight clips

Automatic 15-second vertical highlight clips; 850nm infrared night monitoring that never disturbs the fish.

💧

Full water sensing

Dual temperature cross-check + TDS + pH + dual level + leak detection; AI turbidity / green-water / oil-film assessment.

Progress · 2026-09

Transparent R&D progress

The percentage is assessed from completed milestones across the whole development plan, and updated as development moves forward.

✅ Completed

  • PRD, hardware BOM and a 16-week MVP plan (486 person-days) all frozen
  • Vision algorithm smoke test passed end-to-end: synthetic-data YOLO11n mAP50 = 0.995, counting error ≤±1
  • LLM care assistant prototype verified offline: 35-entry knowledge RAG + medication-dosage guardrail blocking
  • AI training platform integration complete with 49 tests green

🚧 In progress / next

  • Real-data M0 vision validation experiment and Gate decision
  • Team up and start MVP development and assembly
  • Imaging prototype (IMX415 + CPL polarizer) build
Technology

Key technology

EditionsBasic: ESP32-S3 (¥299–499) / AI: RK3588S 6-TOPS NPU (¥999–1699)
OpticsIMX415 8MP + CPL polarizer + CRI>90 pulsed white light + IRC dual-filter night switching
AlgorithmsYOLO11 + ByteTrack detection & tracking, RKNN INT8 quantized on-device inference
SensorsDual DS18B20 temperature cross-check, TDS, pH, dual liquid level, leak detection
SafetyIndependent safety MCU: separate over-temperature breaker + 15-min water-line fuse
AssistantLLM care assistant: knowledge RAG + live tank data + medication-dosage guardrails
SubscriptionFree: 7-day cloud storage / daily report; Pro ¥9.9–19.9/month (30-day storage / health report / unlimited assistant)
How to join

Back the AI Smart Aquarium crowdfunding

🚀

Early access · Founder pricing

Crowdfunding backers get priority access to engineering samples and founder pricing, plus first updates on mass-production schedules.

🤝

Co-development · Custom

Enterprises can co-develop on any active project: custom sensors, enclosures, apps and cloud platforms — sharing the R&D outcome.

📈

Channel · Industrial partners

Distributors, system integrators and industrial investors are invited to participate via crowdfunding or strategic partnership.

What is the difference between the Basic and AI editions?
The Basic edition (ESP32-S3) focuses on sensing and core monitoring at ¥299–499; the AI edition adds the RK3588S 6-TOPS NPU for on-device vision (counting / dead-fish detection / activity / clips) at ¥999–1699.
What does '90 days without losing a fish' mean?
It is the product's north-star metric: a self-reported 90-day fish survival rate ≥70% for beginners (industry baseline 40–50%), approached through AI vision alerts + the care assistant + water-change safety constraints. The project is at algorithm validation / pre-MVP stage with transparent progress updates.
Risks & Challenges

We put risks on the table

Crowdfunding means sharing development uncertainty with us — please read these before backing.

  • Algorithm validation: current metrics come from synthetic-data smoke tests; real-scene accuracy (glare, murky water, dense planting) requires M0 validation — we adjust optics before mass production if needed.
  • Delivery timeline: the AI edition integrates structure, optics, algorithms and the safety chain; the 16-week MVP plan may slip, with weekly progress updates on the site.
  • Living animals: fish losses also depend on individual health and shipping — the product reduces risk but cannot guarantee survival.
  • Safe usage: heater and water-line safety are guarded by an independent MCU; users should still inspect equipment as instructed.

Got an idea? A machine to make smart?

From a brand-new AI device to retrofitting existing equipment — talk directly with our engineers. Bilingual service in English & Chinese.

📞 Phone / WeChat: 19040667970
✉️ buy@allinai5.com Mon–Fri 9:00 – 19:00 (UTC+8) · Shenzhen, China · Serving worldwide