WildGuard is a solar-powered, AI-driven robotics system that continuously monitors forest roads to detect wildlife, deploy escalating deterrents, and issue real-time alerts to prevent human-animal conflicts
A solar-powered camera and thermal node that detects wildfire smoke and ignitions entirely on an Arduino UNO Q, using integer-only SuperFloat CNNs. It alerts over whatever link survives, and no image ever leaves the board.
BRASA is an open-source, low-cost node that watches a home after its residents evacuate, uses on-device AI to detect a growing ember ignition, sprays only the affected zone, verifies the result thermally, and reports to responders over an encrypted offline LoRa mesh.
REMA is an offline Edge AI assistant designed to support people during emergencies and infrastructure failures. It identifies available resources, translates them into practical capabilities, and helps users prioritize those capabilities according to their immediate needs. Unlike a generic chatbot or survival guide, REMA uses a local resource capability map and validated safety rules to provide recommendations based on what is actually available. The system can update its recommendations when resources are added, removed, or depleted. REMA is designed to work without internet or cloud connectivity, making it useful when critical infrastructure is unavailable. When infrastructure fails, use what remains.
A low-cost, offline edge AI kit that scans homes for earthquake hazards like unanchored furniture and heavy objects above beds, keeping every photo private on the device.
A smart gym vending system with custom software, cashless payments, real-time inventory tracking, remote monitoring, and sales analytics for fitness centers.
digitalmonk321
Edge wildfire detection and evidence fusion: a detector on any camera, as small as an ESP32, one integrator to weigh their reports. Kernwerk.
a smart speaker turned into an AI assistant.
ShelterSense is an edge AI system for predicting developing environmental and infrastructure failures in emergency shelters through a network of environmental, air-quality, and occupancy sensors.
Gusty is a mesh network for reporting of thunderstorm gust fronts through a cooperating network of automated weather stations, Edge AI, and a LoRa based MeshCore sensor network
Edge wildfire detection and evidence fusion: a detector on any camera, as small as an ESP32, one integrator to weigh their reports. Kernwerk.
Ultra-Early Off-Grid Wildfire & Ember Acoustic-Visual Detection Node
m_w
Autonomous edge AI nodes that detect disasters like fire and flooding, make local decisions, and warn people in real time — with zero cloud dependency. Built on the Arduino UNO Q with Edge Impulse models.
Resilience Deck is an open-source, offline-first field computer designed to help communities diagnose, understand, and restore critical systems when normal infrastructure is unavailable.
A 7-slot medication dispenser with motor control and camera detection of pill intake. Notifies users on an app when to take their medication and whether they missed a dose.
![PlantVillage Disease Classification [AKIDA HARDWARE]](/api/image-proxy?url=https%3A%2F%2Fdeveco.nyc3.digitaloceanspaces.com%2Fcover%2F695f0755-bd94-4bb4-b364-9640e4a91653.png&w=600)
This project walks through evaluating a PlantVillage AkidaNet model on Akida and, if a hardware device is available, benchmarking its latency and power on a physical AKD1500. It's the hardware companion to the PlantVillage Disease Classification software project. If you haven't seen that yet, it walks through training, quantizing, and converting the model this project benchmarks.
martin-BrainChip
This year I decided to create my own add-on badge for Open Sauce, which resulted in this fun LED flasher badge based on a simple 555 timer circuit.
knitronics
This project walks through the complete pipeline to train, quantize, convert, and benchmark an AkidaNet model on the PlantVillage dataset for Akida 1 hardware. PlantVillage contains 54,303 images of healthy and diseased plant leaves across 38 categories (14 crop species × multiple disease types plus healthy variants). The task is a 38-class image classification problem: given a 224×224 RGB image of a leaf, identify the crop species and disease (or healthy state).
martin-BrainChip
I'm building an application for elderly care
A mod of the Hack Pack Omnibot Forklift adding precise maneuvers, a robotic gripper, a web control interface, remote camera streaming, and much more.
HeavyFalcon678