NASA HUNCH Project 2026-27

Autonomous AI Plant Growth Lab

An intelligent, enclosed autonomous growth chamber designed for lunar & space habitats. Featuring 24/7 computer vision stress diagnosis, micro-dosing irrigation, and autonomous environmental control.

View System Dashboard Hardware & CAD Specs

Problem Statement

Strict Spatial & Resource Limits: Current space plant enclosures operate under tight space and resource constraints.
Manual Monitoring & Care: Astronauts must manually water and inspect crops, draining crew time.
High Crew Time Drain: This labor-intensive process consumes valuable crew hours that could be spent on primary mission tasks.

Project Requirements

24/7 continuous environmental sensor monitoring.
Moisture sensors detect individual plant water content in real-time.
Automated micro-dosing water delivery matching specific plant needs.
Enclosed growth habitat with multi-crop capacity (3 or more plants).

System Capabilities

Integrating durable hardware engineering with an autonomous data-driven SCADA software stack.

Physical Prototype Capabilities

Hardware Specs

Designed for confined space habitat integration with structural wood framing, transparent acrylic viewing panels, and flexible sensor mounting points.

Dimensions	24" W × 24" D × 25" H
Structure	Wood framing + Plexiglass viewing panels
Dosing Pump	Jecod DP-3 Auto Dosing Pump System
Lighting	Viparspectra LED Growth Light Array
Moisture Sensing	SEN0114 Soil Moisture Sensor Nodes
Maintenance	Hinged back door for rapid maintenance access

Code & AI Pipeline Capabilities

Software Stack

An end-to-end data processing pipeline (SCADA) that evaluates sensor telemetry and applies computer vision models to diagnose crop health.

Edge Compute	Raspberry Pi Data Gathering Engine
Mobile / Web UI	Custom Monitoring Application Architecture
Container AI	Anomaly detection for temperature, air & moisture
Plant Health AI	Computer vision trained on leaf coloration & growth patterns
Restorative Action	Automated micro-dosing and parameter correction