Skip to content

Repository files navigation

Drone-Based Volume Calibration for OptiTrack

Automated drone-based calibration for large OptiTrack camera volumes.

Project Description

This project gives OptiTrack operators a way to calibrate large camera volumes — sports fields, multi-story buildings, and other spaces too big or too tall for a person to wand by hand — by sending a drone through the space instead.

Manual calibration of a large volume is slow, hard to reproduce, and sometimes physically impossible. Walking a calibration wand across a 100-meter field or up several stories of scaffolding is exhausting at best and unsafe at worst, and any inconsistency in coverage shows up as worse tracking accuracy.

By having a drone fly a defined trajectory while carrying a calibration rod, the same volume can be calibrated repeatably, with full coverage, without putting a person in the space. The flight path can be saved and re-flown the next time the volume needs to be re-checked.

Key Features

  • Qt-based waypoint editor. Operators sketch and edit the drone's flight path through the volume in a desktop UI — no scripting, no flight-controller knowledge required.
  • Record and replay calibration runs. Once a trajectory works for a given room or field, it can be saved and re-flown on demand whenever the volume needs to be re-calibrated.
  • Calibration-rod handling with Motive integration. The drone carries the calibration rod through the volume and coordinates start/stop with OptiTrack Motive, so calibration data lands in the same workflow operators already use.
  • Onboard SLAM mapping. The drone builds a live map of the space as it flies, so operators can see the volume the drone has covered and confirm the calibration trajectory matches the real environment.

Qt waypoint editor showing a user-defined drone flight path through a calibration volume Figure 1 — The Qt waypoint editor: operators place and edit waypoints to define the drone's calibration trajectory.

SLAM map generated by the drone during a calibration flight Figure 2 — Live SLAM map produced by the drone as it flies, showing the volume covered during a calibration run.

<script type="module" src="https://unpkg.com/@google/model-viewer@3.5.0/dist/model-viewer.min.js"></script>

Interactive 3D model of the calibration drone — view at the GitHub Pages site to rotate and inspect.

Figure 3 — Interactive 3D model of the calibration drone (drag to rotate; scroll to zoom). Visible on the GitHub Pages version of this README.

How to Access or Try It

This is a research and development project; there is no installable build yet. To see how it works:

System requirements:

  • OptiTrack Motive (for the camera volume being calibrated)
  • A VOXL-based drone (the on-drone software lives in voxl-programs/)
  • Qt, to run the desktop control UI

Team

This is an Oregon State University capstone project, developed in collaboration with NaturalPoint Inc. DBA OptiTrack.

For questions or feedback, email any of the team members above.

License

All non-third-party code in this repository is jointly owned by the team listed above and NaturalPoint Inc. DBA OptiTrack. See LICENSE.txt in the licenses/ folder for full terms; third-party licenses are in the same folder, named accordingly.

About

Project for using a drone to calibrate a large OptiTrack system.

Resources

Contributing

Stars

0 stars

Watchers

0 watching

Forks

Releases

Packages

Contributors

Languages