Skip to content

index

Jump to bottom
Eric Busboom edited this page Aug 31, 2026 · 2 revisions

AprilCam v2 Docs

Navigation map for the AprilCam v2 subsystem — daemon, MCP server, robot client library, and the wire protocol.

AprilCam v2 Documentation

AprilCam gives AI agents and robot programs a shared, real-time picture of a robotics playfield — AprilTag/ArUco positions, orientation, velocity, and homography — served by one daemon that owns the cameras and does all the vision. This wiki documents AprilCam v2, the ground-up rebuild: a new gRPC protocol, a world-coordinate-first tag pipeline, and thin clients that never touch pixels. Pick your entry point:

Start here

  • Overview — what AprilCam v2 is, how the daemon / MCP server / client library / viewer fit together, the world-coordinate conventions, and the install tiers.

Use it

  • Using the MCP Server — for AI agents. Run aprilcam mcp, follow the golden path, and drive the full tool catalog — perception, configuration and calibration, mobile tags, and shared annotations — over the Model Context Protocol.
  • Robot Direct API (the library) — the aprilcam.client Python library for robot control loops: discover and connect to the daemon, read tag positions at loop rate, subscribe to tag and image streams, and draw shared annotations over gRPC.
  • Operating the Daemon — install, run, configure, and troubleshoot the daemon: the aprilcam daemon lifecycle, the config cascade and directory layout, mDNS advertisement, systemd, and the troubleshooting playbook.

Build against it

  • Daemon Wire Protocol — the AprilCamV2 gRPC control service, the length-prefixed protobuf stream sockets, and the core message schemas. Read this to implement a client in another language.

Background

Maintainers: this wiki is published to the hub automatically on push to docs/wiki/** on master. See /AGENTS.md and the source map in _subsystem.yml for how to keep these pages in sync with the code.

Clone this wiki locally