Generic Sensor Feed → TAK (template)¶

A minimal, wired-up skeleton for polling an arbitrary REST API on an interval and publishing the results as CoT. Rename the tab, edit four nodes, deploy.

This is the starting point for CAD-dispatch feeds, river-gauge feeds, lightning-strike feeds, vehicle telemetry, and any other REST-polled data source not covered by a dedicated library flow.

Read the Flows overview first

Import the GeoJSON to CoT subflow before importing this flow.

Shape¶

[ Every 60s ] → [ Feed URL ] → [ Fetch feed ] → [ Map to GeoJSON ] → [ GeoJSON → CoT ] → [ TAK Server ]

All five wires are already connected. You only edit nodes marked EDIT ME and the subflow instance config.

Install¶

Before you start¶

Flows overview prerequisites done.
GeoJSON to CoT subflow imported.
Service account created.

1. Import¶

Node-RED → ☰ → Import → Local → fasttak → generic-sensor → Import.

2. Configure TLS¶

☰ → Configuration nodes → Generic Sensor Server TLS → "Use key or certificate from local file" → replace {svc_user} → Update.

3. Fill in the four EDIT ME nodes¶

a. `Every 60s — EDIT ME` (inject)¶

Set repeat to your desired poll interval in seconds. Respect upstream rate limits.

b. `Feed URL — EDIT ME` (change)¶

Set msg.url to your REST endpoint. If you need query parameters, headers, or auth, add them to the Fetch feed http-request node or build them in an intermediate change/function node.

c. `Map to GeoJSON — EDIT ME` (function)¶

This is where most of your work happens. Transform msg.payload (your API's response body) into a GeoJSON FeatureCollection. The required shape:

msg.payload = {
    type: 'FeatureCollection',
    features: [
        {
            type: 'Feature',
            id: stableIdForEntity,
            geometry: { type: 'Point', coordinates: [lon, lat] },
            properties: {
                callsign: '...',
                remarks: '...'
                // Optional: altitude, speed, course for air/sea tracks
            }
        }
    ]
};
return msg;

Tip: use node.warn(JSON.stringify(msg.payload, null, 2)) during development to inspect the shape of your API response.

d. `GeoJSON → CoT` (subflow instance)¶

Double-click and set:

COT_TYPE — which TAK marker symbol to use. See docs/reference/cot-types.md for common types, or the full catalog.
UID_PREFIX — unique per flow (so IDs don't collide with other flows).
STALE_SECONDS — must be greater than your poll interval.

Example: CAD dispatch feed¶

Given an API that returns:

{
  "incidents": [
    { "id": "INC-123", "type": "Structure Fire", "lat": 39.7, "lon": -104.99, "units": "E5,T2", "address": "1234 Main St" }
  ]
}

Mapping function:

const raw = msg.payload;
const features = (raw.incidents || []).map(inc => ({
    type: 'Feature',
    id: inc.id,
    geometry: { type: 'Point', coordinates: [inc.lon, inc.lat] },
    properties: {
        callsign: inc.type,
        remarks: `${inc.type} @ ${inc.address} | Units: ${inc.units}`
    }
}));
msg.payload = { type: 'FeatureCollection', features };
return msg;

Subflow config: COT_TYPE=a-f-G-E, UID_PREFIX=cad, STALE_SECONDS=600.

Files¶

nodered/flows-library/generic-sensor.json