Ericsson Logo



Selected track
Track ID--
Track type--
Source--

First seen--
Last seen--
Track age--

Speed--
Altitude--
Location--
Current time--

Classification--

Track length--
Link ID--
Points--
๐Ÿ‘

3D MAP

Wheel to zoom, Left click and drag to orbit, or keys WASD + keys QZEC to move.

3D Map visualization placeholder.

Sensor tracks System tracks Traffic Ground truth

Metrics

Live Plots

When any plot data arrives on the MQTT topic EricssonGarageLund/ISAC/live_plot_metrics/garage/# you will see it here.

Sensor Min Error (m) Tracked
101_311 - -
101_331 - -
311_101 - -
311_331 - -
331_101 - -
331_311 - -
System - -
No data received yet

Archive

System Status

GUI Uptime: 00:00:00
-
When any plot data arrives on the MQTT topic EricssonGarageLund/ISAC/live_plot/garage/# you will see it here.

System Configuration

Here's the config tab.

System Installation

Send the message "ping" to the ISAC mqtt topic EricssonGarageLund/ISAC/system_daemon/install

Your device does not support Augmented Reality

The device you're using right now to read this does not seem to be able to handle Augmented Reality content properly.

Either you'll need a device with a front-facing camera and orientation sensors, such as an iPad or Android tablet. A mobile phone could work as well, if you're ok with a small screen.
You can also use a Virtual Reality device that supports WebXR such as Meta Quest 3.

Your device needs to have a front-facing camera and orientation sensors, to be able to handle Augmented Reality. With such a device, this tab will show you overlaid markers on top of the camera view on your device, showing the location of tracked objects combined with the live real-world footage from your device's camera.

TOPOLOGY

Work in progress, putting height data into this 3d map. Wheel to zoom, Left click and drag to orbit, or keys WASD + keys QZEC to move.

Map overview

Latest actions

    System connections

    SKYKEEPER
    SIGMA
    FIVE EYES
    IKE
    9AIR-C4I
    TACS
    LATTICE

    Selected object

    Incident โ€”
    Validated by โ€”
    Validated at โ€”
    Speed โ€”
    Altitude โ€”
    Location โ€”
    Tracking accuracy โ€”
    Classification
    Intent
    Note

    Detections

      Tracked objects

      ID Class Time Location Speed Alt Hdg
      Ericsson Sensing (alpha, in development)
      User name
      Observed Area

      Ericsson Sensing Developer Applications

      Launches apps on test.isac.egl.seld.rnd.ericsson.se/apps/ (via apps.php proxy). Enter your RADIANT Apps password (not saved) to list, start and stop apps for your username.
      AppDescriptionStatusAction

      Ericsson Sensing GUI Help

      Ericsson Sensing, enabled by ISAC

      Sensing
      GUI

      Real-time geospatial intelligence platform for drone tracking, sensor fusion, and radio node monitoring.

      Continuous evolution - now split into five apps

      Release Notes Version History

      Audience Five App Versions

      Ericsson Sensing Demo
      Command and Control center โ€” multi-client collaboration setup for coordinating operators in real time.
      Ericsson Sensing (alpha, in development)
      Operator mode โ€” EDS-styled Command and Control with the same C2 capabilities and Ericsson Design System look and feel.
      Ericsson Sensing Viewer
      Guest / observer mode โ€” for stakeholders interested in technical details, without system control access.
      Ericsson Sensing GUI
      Administrator mode โ€” full control of the Ericsson Sensing ambulating system, node management, and sensor configuration.
      Ericsson Sensing dev
      Developer mode โ€” advanced tools for improving and debugging detection algorithms, with access to all experimental tabs.

      Overview What is Ericsson Sensing Demo?

      • Command and Control center โ€” multi-client collaboration setup for coordinating operators in real time
      • 2D / 3D map panel โ€” switchable between Leaflet (OpenStreetMap or Google Maps) and custom WebGL 3D view with optional Google 3D tiles
      • Zone drawing tools โ€” draw perimeters, no-go zones, and ignore zones directly on the map
      • Real-time detections feed โ€” timestamped detections for perimeter breach, swarm pattern, loitering, and rapid approach events
      • System connections panel โ€” live link-status for all connected external systems

      Overview What is Ericsson Sensing Demo? (cont.)

      • Selected object inspector โ€” speed, altitude, classification, and tracking accuracy for any tracked object
      • Collaboration groups โ€” named MQTT groups; multiple operators share one session with HUD notifications on join and partial username privacy
      • Remote scenario loading โ€” loading a scenario broadcasts it to all group members and synchronizes the timeline replay position automatically
      • Auto-adjusted latency โ€” GUI detects Sensing system latency at runtime and shifts the display timeline accordingly (capped at โˆ’30 s)
      • Scenario loader โ€” switch between live MQTT view and pre-loaded archive test scenarios
      ๐ŸŒ
      Technology Stack
      Any modern web browser ยท Ericsson Design System ยท Leaflet.js ยท MQTT.js ยท JSZip ยท WebGL/WebGL2

      Unit tests ยท Test suites ยท Playwright e2e

      C2 Demo Five Stories for the C2 View

      The C2 view supports a story-driven demonstration designed for security analyst audiences. Pre-loaded scenarios allow a presenter to walk through all five stories without a live sensor feed. All state is synchronized across connected clients in real time.

      1. Object Detection
      Monitor a pre-defined area. New objects appear on the map and trigger notifications.
      2. Tracking & Classification
      Turn raw detections into a prioritized threat picture with type and intent classification.
      3. Recording, Validating & Alerting
      Record incidents, validate tracks, add notes, and alert commanders.
      4. Geofencing
      Draw protected and ignored zones with intent-based alert rules.
      5. Data Export & Integration
      Connect to external C2 systems and export operational data.

      Story 1 Object Detection

      Security monitors a pre-defined area. When an object is detected, it appears on the map and triggers a notification. The analyst decides whether to track or ignore it.

      Demo checklist

      1. Load a scenario from the Live view dropdown (or use live MQTT)
      2. Point out unclassified objects on the map โ€” they show a pulsating white ring
      3. Show the Detections panel (right side) โ€” new objects appear with color-coded urgency dots
      4. Click an object on the map โ€” a Track / Ignore popup appears
      5. Click Track to promote the object, or click the Track link in the Detections list
      6. Show that tracked objects move from Detections to the Tracked Objects table
      7. Click Ignore on a non-threatening object โ€” it is dismissed with an audit log entry

      Story 2 Tracking & Classification

      Security teams turn raw detections into a prioritized threat picture by classifying each object's type and intent.

      Demo checklist

      1. Select a tracked object โ€” the ID card (Selected Object panel) shows speed, altitude, location, heading
      2. Point out the Suggested classification hint below the dropdowns โ€” the system proposes a type based on speed and altitude
      3. Click Accept to apply the suggestion, or manually choose a type from the Classification dropdown (Drone, Helicopter, Plane, Car, Bus, Tram)
      4. Set the Intent dropdown (Friendly, Suspicious, Hostile) โ€” the map marker and table row color update immediately
      5. Show the map icons โ€” classified objects display an EDS icon (drone, car, plane, etc.) on a semi-transparent enlarged marker
      6. Use the filter tabs (All / Friendly / Threats) in the Tracked Objects table to show how the threat picture is prioritized
      7. Point out the color scheme โ€” green = friendly, amber = suspicious, red = hostile, grey = unclassified

      Story 3 Recording, Validating & Alerting

      Security teams record and validate tracked objects, enrich them with notes, and take action on prioritized threats.

      Demo checklist

      1. Select a tracked object and show the Tracking accuracy percentage and Validated by field (system or analyst)
      2. Type a note in the Note textarea โ€” explain that notes are synced across all connected C2 clients
      3. Click Alert โ€” a commander notification is logged and the button briefly shows "Alert sent!"
      4. Click Disable SIM โ€” show the planned-capability placeholder dialog (future integration)
      5. Click Center on map โ€” the map pans and zooms to the selected object
      6. Click View action log โ€” the log dialog opens, filtered to show only events for this specific object
      7. Show the Latest Actions panel (bottom left) โ€” all operator actions appear as a timestamped audit trail

      Story 4 Geofencing

      Security teams define protected zones and receive real-time alerts when objects with matching intents violate them, or mark areas to ignore to reduce noise.

      Demo checklist

      1. Click Draw new perimeter on the map โ€” click to place vertices, then click Done
      2. In the dialog, choose a preset (No-go zone, Area of interest, Flight corridor, or Ignore zone)
      3. Set the alert level (Red / Orange / Yellow / Blue / Green) and give the zone a name
      4. Configure Alert rules โ€” check which intents trigger alerts (Unclassified, Suspicious, Hostile, Friendly)
      5. Optionally set altitude constraints (min / max) for a No-fly zone
      6. Show the perimeter rendered in both the 2D and 3D map views
      7. Demonstrate a breach โ€” when a matching object enters the zone, a detection is automatically created
      8. Show that an Ignore zone (green / alertLevel 0) suppresses all detections within it

      Story 5 Data Export & Integration

      Security teams stream detections, classifications, and incidents to external systems for a unified operational picture.

      Demo checklist

      1. Show the System Connections panel (bottom left) with on/off toggles for 7 external systems
      2. Toggle a system on or off โ€” point out the audit trail entry in Latest Actions
      3. Click the three-dots menu (โ‹ฎ) on the System Connections panel heading
      4. Select Show information โ€” walk through the system info presentation deck (SKYKEEPER, SIGMA, etc.)
      5. Select Data shared โ€” show which data categories are shared with connected systems
      6. Select Connection status โ€” show per-system connection state and timestamps
      7. Select Export data โ€” download the full C2 state as a JSON file
      8. Mention that API-based streaming to external C2 platforms is a planned capability

      Overview What is Ericsson Sensing GUI/Viewer/dev?

      • Administration โ€” full control of the Ericsson Sensing ambulating system, node management, and sensor configuration
      • Research framework โ€” a wide range of visualizations for research, debugging and implementation
      • Live MQTT streaming โ€” ingests position data from drones, sensors, and radio nodes in real-time
      • Multi-format parsing โ€” handles Drone OSD, Fusion Server, OwnTracks, Trafiklab, and Ericsson proprietary formats
      • Archive playback โ€” replay historical sessions from compressed log files
      • Zero build step โ€” pure vanilla JS, runs directly in any browser
      • Desktop, tablet or mobile โ€” optimized for any screen size
      ๐ŸŒ
      Technology Stack
      Any modern web browser ยท Ericsson Design System ยท Leaflet.js ยท MQTT.js ยท JSZip ยท WebGL/WebGL2 ยท A-Frame.js + AR.js

      Unit tests ยท Test suites

      Quality Assurance Test Coverage

      3662 unit tests
      104 test suites
      2 test runners
      211 e2e tests
      3 browsers
      630 total runs
      Core Data
      TrackStore SoA, interpolation, extrapolation, registry, Node class
      Protocol Parsing
      Drone OSD, Fusion, OwnTracks, Trafiklab, Ericsson, golden-ref formats
      UI Logic
      Timeline, Maps, Maps POI, 3D map, Settings, Archive, AR, Topo, Help, Docs, drag divider, config handler, developer mode, MQTT log replay, Track Detail Panel, Google Maps 2D/3D, C2 lobby, C2 detections, C2 Guru, C2 syscon, C2 EDS, C2 EDS syscon, C2 tracked objects, C2 selected object, C2 alert, C2 values, load-order verification

      Jest unit tests ยท Async integration tests included ยท Playwright e2e (chromium, firefox, webkit)

      Current Features

      Production tabs visible to all users โ€” no developer mode required

      Tab 1 Maps

      • Resizable grid layout โ€” multiple simultaneous Leaflet map panels with drag dividers
      • Canvas overlays โ€” custom WebGL-powered sensor track and drone trail rendering
      • Live MQTT ingestion โ€” real-time position updates with configurable trail window (default 10 s)
      • Track filter cache โ€” rebuilt every 1.28 s for smooth 60 fps animation
      • Ground truth mode โ€” display full historical tracks alongside live data
      ๐Ÿ“
      Supported Track Types
      Drone OSD ยท Ericsson Fusion ยท OwnTracks ยท Trafiklab transit ยท Sensor targets ยท Radio nodes with PCI shapes

      Tab 2 Archive

      • Historical playback โ€” load compressed MQTT log files (.zip / .mqttlog) and replay at configurable speed
      • JSZip decompression โ€” client-side, no server round-trip required
      • Syntax-highlighted inspection โ€” JSON messages rendered with truncation for large payloads (>20 KB)
      • Timeline integration โ€” scrub through archive sessions directly on the timeline bar
      • Memory-safe โ€” tab content cleared on navigation to prevent layout reflow overhead

      Tab 3 Status

      • Live MQTT topic explorer โ€” hierarchical tree view of all active topics
      • Message inspector โ€” click any topic to view latest payload
      • Topic statistics โ€” message rates, last-seen timestamps
      • Auto-clearing on tab exit โ€” tree rebuilds from live data on re-entry
      ๐Ÿ“ก
      MQTT Connectivity
      WebSocket connection ยท Configurable broker endpoint ยท Reconnect handling ยท Multi-topic subscription management

      More Tabs Docs & Settings

      ๐Ÿ“„
      Docs Tab
      In-app documentation โ€” Open-Source Software Licenses. Always up-to-date alongside the code.
      โš™๏ธ
      Settings Tab
      MQTT broker setupTrail ยท Development mode enable ยท Trail window slider (localStorage persisted)

      Future Features

      Advanced tabs unlocked in developer mode โ€” coming to production soon

      Dev Feature 3D Map dev

      • Custom WebGL engine โ€” no Three.js, matrix-based camera transforms from scratch
      • Google Maps 3D tile support โ€” optional photorealistic 3D terrain and buildings with automatic fallback
      • WASD + QZ keyboard navigation โ€” fly-through orbit controls with scroll zoom and camera-following drone view
      • Live + archived tracks โ€” drone paths rendered alongside real-time positions
      • Future position projection โ€” extrapolated trajectories with configurable lookahead
      • Optimised geometry โ€” thick-line shader, unit sphere template, inline Mercator projection
      ๐Ÿ—บ
      View Controls
      Show/hide sensing targets ยท system tracks ยท traffic ยท future data ยท obsolete tracks ยท extrapolation toggle โ€” all persisted to localStorage

      Dev Feature Augmented Reality dev

      • A-Frame v1.4 + AR.js โ€” WebXR-based AR overlay, lazy-loaded on demand
      • Live track overlay โ€” drone and sensor positions projected into camera view
      • Mobile-first โ€” designed for tablet and smartphone field use
      • Zero-install โ€” runs entirely in the browser, no native app required
      ๐Ÿ“ฑ
      Use Case
      Point a tablet at the sky โ€” see real-time drone tracks overlaid on the live camera feed. Designed for field operators and site surveys.

      Dev Feature Track Detail Panel dev

      • Selected-track side pane โ€” dedicated panel for in-depth analysis of any selected track
      • Real-time graphs โ€” speed, heading, altitude, and classification confidence visualized over time
      • Delta-t history โ€” time-between-messages graph for debugging sensor update frequency
      • Interactive charts โ€” hoverable and clickable data points for precise value inspection
      • Drone classification โ€” detailed breakdown of detection algorithm confidence scores
      ๐Ÿ“ˆ
      Use Case
      Algorithm developers can select any track on the map and immediately see its behavioral signature โ€” perfect for validating detection logic, tuning classification models, and debugging sensor noise.

      Dev Features Metrics ยท Config ยท Installation dev

      ๐Ÿ“Š
      Metrics
      Plotly-based live plots โ€” MQTT throughput, track density, latency histograms. Auto-resizes when tab is brought into focus.
      ๐Ÿ”ง
      Config
      Runtime configuration editor โ€” broker endpoints, topic filters, display thresholds. Changes applied without page reload.
      ๐Ÿ›ฐ
      Installation
      Radio node deployment view โ€” manage sensor grid topology, link states, and coverage maps for Ericsson Sensing infrastructure.

      Launch Testing Strategy

      Unit tests
      complete
      E2E automation
      suite ready
      User testing
      16 Mar 2026
      Continuous
      updates
      Unit & Integration
      Test suites and test files covering all parsers, data structures, UI logic, and async MQTT flows. Jest runner with full pass/fail reporting.
      End-to-End Automation
      Playwright test suite in test_automation/ covers full browser workflows โ€” tab navigation, archive playback, MQTT connection lifecycle.

      File Operations

      Drag and drop files on the timeline at the bottom of the MAPS tab or the 3D MAP tab.

      Accepted file types are:
      * View settings (typically named isac_gui_settings.json),
      * MQTT log files with Ground Truth, System or Sensor tracks (typically named *.mqttlog).

      You can drop multiple files at the same time, or even zip files with multiple files inside. Press the download icon at the upper right corner of the timeline in the MAPS tab or the 3D MAP tab to download a zip file with everything needed to recreate the current GUI state: MQTT log files, isac_gui_settings.json

      Keyboard shortcuts

      Space to toggle timeline pause.

      F4 to enter deep sleep.

      Hold 8 / 9 with a track selected to shift its time offset earlier / later (50 ms per frame). The current offset is shown in a HUD overlay.

      Navigate the 3D MAP:
      W - forward
      S - backward
      A - left
      D - right
      Q - up
      Z - down
      E - raise ground
      C - lower ground
      Click and drag with the mouse

      Sensing GUI

      Real-time geospatial intelligence.
      Continuous evolution.

      Thanks for watching

      Settings

      MQTT Configuration

      URL must start with ws:// or wss://

      User Role

      Performance

      Trail window: 20 s
      Seconds of trail shown around the current time when a selection is active. Lower values may improve GUI performance.
      Latency offset: -1.00 s
      How far behind "now" the visualization lags (seconds). When dynamic adjustment is on, this adapts automatically to measured MQTT latency.
      Memory limit: 4.0 GB
      By limiting the amount of RAM used for storing the history of live MQTT tracks, the GUI will forget the oldest tracks to be able to run 24/7.
      This allocates ~1.5GB of RAM per hour, as this data is quite verbose. The GUI's internal storage is much more efficient.

      Imported data files

      Drag and drop any mqttlog file(s) here - or directly onto the timeline in the maps tab
      These are the files imported into the Ericsson Sensing GUI:

      MQTT subscriptions

      Topics the app subscribes to for live track data. A checked checkbox means the topic is currently active with the broker. Remove user-added topics with the trashcan icon.
      Any data arriving on your added MQTT topic will feed into the timeline and will appear on maps.