Bee Edge AI

Edge Platform for Custom Workloads

Building

At its core, the Bee was designed with a modular, plug-in architecture, making it fast and easy to deploy workloads. Starting today, we’re unlocking three powerful capabilities for developers.

1. Run Custom Edge Detections

You can now deploy custom edge detections on the Bee device—models that run alongside (but do not replace) the native Bee Map AI stack. This lets you extend the Bee’s detection capabilities for your own needs, such as identifying specific utility markers, custom signage, or industry-specific infrastructure.

The core object detection model used for map building is maintained and updated by Bee Maps. However, your edge detections can operate in parallel on the device—filtering, tagging, or flagging new visual classes relevant to your operations.

Deploy models via OTA (over-the-air) using Bee Maps’ geo-targeted infrastructure. You can push to specific devices by region—country, state, or metro—allowing you to target detections based on where and what you need to capture.

The Bee runs a high-efficiency, edge-optimized YOLO model in real time on every frame from its 12.3MP camera. With 5.1 TOPS of compute on board, it delivers low-latency inference even in fully offline or low-bandwidth environments.

How to Deploy a Custom Model

• We collaborate with partners to train objects.

• Your model is deployed to a small subset of Bee devices to validate accuracy and recall.

• Once approved, it’s rolled out to your designated geographies via an OTA update to Bee devices.

2. Change Detection

The Bee’s Map AI continuously detects and maps critical road features, including speed limit signs, stop signs, road width, and other street-level attributes. Additionally, when deploying custom object detections—such as utility infrastructure—you can leverage Change Detection to efficiently identify and upload only the changes (deltas), significantly optimizing data transmission.

Real-Time Detections	Compare Against
Bee - Native Mapping Detections (eg speed signs, traffic lights, etc.)	Your own dataset for these objects
Your Custom Object Detections	Your own dataset for these objects

You can track changes to specific infrastructure objects—like speed limit signs—by uploading a reference dataset in GeoJSON format. Each object should conform to our feature spec, for example:

{
  "id": "SLS1553923",
  "objectClass": "speed-limit",
  "image-detection": "https://your-bucket-name.s3.beemaps.com/path/to/image.jpg",
  "attributes": {
        "speedLimit": 50,
        "unit": "mph",
        "regulatory": "enforced"
      },
   "position": {
      "lon": -108.123,
      "lat": 34.234,
      "azimuth": 90,
      "height": 0.5,
      "width": 0.3
  },
}

Bee Maps automatically partitions and distributes these objects to devices most likely to encounter them. As Bees detect and upload fresh observations, your GeoJSON file is updated. You can subscribe to real-time changes via API to track updates as they occur.

3. Direct Data Offload to Your Cloud

All data generated from your custom edge detections and change detection workflows—along with optional supporting imagery—is automatically streamed via Bee Connectivity Services from the Bee to your Bee Maps developer account. Bee Connectivity Services leverages the device’s built-in LTE and WiFi offload channels, requiring no additional integration work on your part.

Simply configure your endpoint, and the Bee will immediately begin transmitting structured visual data in real time, ready for ingestion into your systems or workflows.

Contact us ([e-mail]) to become a Bee developer.