TerraGemma: An Offline Edge AI Trail Companion That Tells You to Put Your Phone Away

This is a submission for the Hacktoberfest Open-Source AI Challenge Week 1: Touch Grass

What I Built

Modern hiking and outdoor apps suffer from a tragic paradox: they encourage you to stare down at a glowing screen while walking through ancient forests and alpine peaks. Turn-by-turn navigation chimes buzz on your wrist every 30 seconds. Hikers miss the rustle of leaves, the scent of damp hemlock needles, and the warning shift of mountain winds.

Even worse, the moment you hike a mile past the trailhead, cell towers vanish. The second you enter deep wilderness, every proprietary cloud AI assistant (ChatGPT, Claude, Gemini API) turns into a completely useless spinning loading wheel. If a hiker sprains an ankle, gets stung, or loses daylight, closed AI leaves them stranded.

I built TerraGemma: an offline-first backcountry expedition companion powered by Google's open-weight Gemma 3 running 100% locally on-device.

The Core Philosophy: "Eyes Up, Screen Down"

TerraGemma is engineered around a radical design principle: the screen must be the shortest part of your outdoor experience.

Instead of keeping you glued to a feed or turn-by-turn phone map:

1. NOAA Solar Engine & Naismith Turnaround Math: Calculates exact astronomical sunset, golden hour, and a strict Turnaround Deadline (incorporating elevation gain penalties via Naismith's Rule) so you never get caught on the mountain after dark.
2. "Eyes-Up, Screen-Down" Sensory Navigation Guide: Gemma 3 synthesizes 3 to 4 physical sensory cues (e.g., "Listen for the feeder stream at mile 0.8; feel the wind shift from the western valley as you exit the treeline") so you navigate by natural awareness rather than screen-staring.
3. Screen-Free Audio Briefing ("Listen & Pocket Phone"): Plays a concise audio briefing at the trailhead using the browser's native Web Speech API—no cloud speech APIs, no network needed. You listen once, pocket your device, and start hiking.
4. Offline Wilderness First Responder (WFR) Triage: When emergency strikes with zero cellular bars, Gemma 3 provides instant triage for hypothermia, heat illness, venomous snakebites, anaphylaxis, and sprains, accompanied by international 3-blast whistle rescue signaling protocols.
5. Field Flora, Fauna & Toxic Lookalike Protection: Side-by-side identification of dangerous lookalikes (Poison Ivy vs. Virginia Creeper; Queen Anne's Lace vs. Deadly Poison Hemlock; Black Bear vs. Grizzly) plus mindful "Touch Grass" sensory missions.

Demo

1. Planning with Solar Precision & Safe Turnaround

Selecting a backcountry trail calculates astronomical solar timestamps (Sunrise, Sunset, Civil Dusk) using offline NOAA formulas and applies Naismith’s Rule with elevation gain penalties. A color-coded daylight status badge and a hard Turnaround Time ensure hikers never get caught on the mountain after dark, displayed alongside an interactive OpenStreetMap route.

2. "Eyes-Up, Screen-Down" Audio Briefing

Gemma 3 synthesizes memorable physical sensory anchors (the sound of feeder streams, shifting ridge winds, distinctive boulder cairns) so hikers navigate with their natural senses instead of checking turn-by-turn phone notifications. Clicking *"Listen & Pocket Phone"** activates the offline Web Speech API to read the briefing aloud, allowing the hiker to stow their device and hike hands-free.*

3. Offline Emergency First Responder Triage

In deep backcountry with zero cellular signal, emergency medical decisions cannot wait. Gemma 3 evaluates acute symptoms offline—providing immediate 3-step triage actions, critical contraindications (what NOT to do), evacuation urgency, and international 3-blast whistle rescue signaling protocols.

4. Botanical Safety & Mindful Outdoor Missions

An offline botanical and wildlife field guide comparing dangerous lookalikes side-by-side (Poison Ivy vs. Virginia Creeper, Queen Anne’s Lace vs. Deadly Poison Hemlock, Black Bear vs. Grizzly) alongside mindful "Touch Grass" missions designed to foster genuine connection with the forest.

Here is how TerraGemma works in practice at the trailhead:

1. Expedition Setup & Astronomical Calculations: Select a backcountry trail (or enter custom coordinates). The offline NOAA solar engine calculates sunrise, sunset, and civil dusk. Naismith's rule factors in trail distance and 2,000+ ft of vertical ascent, instantly issuing a color-coded daylight status (🟢 Safe, 🟡 Caution, 🔴 Hazardous) and a hard turnaround timestamp.
2. Interactive OpenStreetMap Route: Waypoints, stream crossings, switchbacks, and high-altitude checkpoints render on an open geospatial topographic map.
3. Synthesize & Pocket Phone: Hit "Synthesize Sensory Field Briefing". Local Gemma 3 generates sensory landmark anchors in milliseconds. Click "Listen & Pocket Phone"—the voice player reads the briefing aloud so you put your phone into your backpack and touch grass.
4. Offline Emergency Triage: If an emergency arises, navigate to the WFR tab to get calm, step-by-step first responder protocols even in airplane mode.

Code

The complete source code is open-source under the MIT License on GitHub:

👉 https://github.com/shivamgravity/terragemma

How I Built It

TerraGemma is built in Python using Streamlit, engineered specifically for 100% offline, zero-cloud execution on consumer hardware:

1. Local Open-Weight AI (Google Gemma 3):
   • gemma3:1b (Primary Edge Model): Ultra-lightweight (~815 MB), running inference in ~500ms on standard laptop CPUs. Perfect for instant trail briefings and field checklists without draining device battery.
   • gemma3:4b (Analytical Model): Deeper contextual reasoning for complex wilderness emergencies and botanical identification.
   • Ollama Engine: Connected via local HTTP API with keep_alive: 60m to ensure warm model weights stay in memory.
   • Resilient Fallback Engine: Built-in heuristic fallbacks ensure zero downtime when deployed for cloud demos.
2. Deterministic NOAA Solar & Naismith Engine:
   • Pure Python mathematical implementation of the NOAA solar algorithm (zenith 90.833° for official sunset, 96° for civil dusk). Zero astronomy C-dependencies.
   • Naismith's Rule with Scarf's equivalence: calculates base pace plus 1 hour per 2,000 ft of vertical ascent and required safety buffers.
3. Offline Web Speech API Audio Synthesizer:
   • Implemented via a self-contained HTML/JS component leveraging window.speechSynthesis. Synthesizes natural speech locally in the user's browser without calling external cloud text-to-speech APIs.
4. Geospatial Mapping (folium & streamlit-folium):
   • OpenStreetMap layers mapping trailheads, water sources, summits, and turnaround checkpoints.

System Architecture

flowchart TD A["Hiker at Backcountry Trailhead (Airplane Mode / Zero Signal)"] --> B["TerraGemma App (Streamlit Local Engine)"] subgraph "Deterministic Outdoor Engines (Zero Dependencies)" B --> C["NOAA Solar Engine (Sunset, Twilight, Golden Hour)"] B --> D["Naismith's Rule Engine (Turnaround Time & Margin)"] B --> E["OpenStreetMap Visualizer (Waypoints & Elevation)"] end subgraph "Open Edge AI Engine" B --> F["Google Gemma 3 (1B / 4B via Local Ollama)"] F --> G["Sensory Landmark Generator ('Eyes-Up Guide')"] F --> H["Wilderness First Responder (WFR Triage Matrix)"] F --> I["Botanical & Toxic Lookalike Field Assistant"] end G --> J["Offline Web Speech API Synthesizer"] J --> K["🔊 Audio Briefing: 'Listen & Pocket Phone'"] K --> L["Hiker Steps Onto Trail with Eyes Up & Mind Clear 🌲"]

Why Does Open Innovation Matter?

In a coffee shop, relying on closed cloud AI APIs (like OpenAI or proprietary cloud endpoints) is convenient. In the wilderness, a closed API is 100% dead.

1. Zero-Connectivity Wilderness Resilience

When you step past the trailhead into a national park, canyon, or alpine pass, cell signal drops to 0 bars. Closed APIs become infinite loading spinners. Open-weight models like Gemma 3 run natively on edge devices. You carry a world-class naturalist and wilderness first responder with you without a single cell tower in sight.

2. Radical Location & Health Privacy

Commercial trail apps frequently package high-precision GPS tracks, departure times, and personal health metrics to sell to ad brokers and data brokers. TerraGemma runs 100% locally: your coordinates, trail notes, and emergency symptoms never leave your device.

3. Zero Running Cost & Unlimited Utility

Closed APIs charge meter fees per thousand tokens and enforce rate limits. On a multi-day backpacking trek, an open-weight model costs $0.00, will never hit an API rate limit, and operates completely indefinitely.

4. Efficient Quantized Edge Footprint

Google's Gemma 3 (1B) requires under 1 GB of memory. It runs on consumer laptops and edge hardware in hundreds of milliseconds without draining critical battery reserves needed for survival navigation.

My Agent Session

I built and verified TerraGemma using DevRelay:

python3 -m py_compile app.py

Syntax OK, all modules verified.

Prize Categories

• Featured Category: Best Use of Gemma ($200 USD): TerraGemma is built entirely around Google's open-weight Gemma 3 (1B and 4B) served locally through Ollama. Gemma powers dynamic sensory landmark generation, structured WFR triage evaluation, and offline botanical identification without any proprietary cloud dependencies.