Calibrex Docs

Calibrex is a ROS-independent calibration evidence framework for robotics and autonomous-driving systems. It turns candidate extrinsics, time offsets, and trajectories into reproducible PASS / WARN / FAIL evidence.

Most calibration tools answer “what transform did the optimizer return?” Calibrex also asks:

• Does it generalize to held-out measurements?
• Can the available data detect a deliberately wrong calibration?
• Which degrees of freedom are weak or unobservable?
• Can the result and every generated report be traced to their inputs?

Start here

1. Run the five-minute quickstart.
2. Choose a path from calibration methods.
3. Reproduce an evaluation from public datasets.
4. Review frame conventions before integrating transforms.

Public real-data benchmark

Five AX=YB methods were recomputed on the ETHZ ASL real robot-arm dataset with one shared 80/20 split.

Method	Rotation holdout RMSE deg ↓	Translation holdout RMSE mm ↓	Known-bad detection fraction ↑	Failure rate	Runtime s
Shah	0.585795	10.0628	1	0.0%	—
Li-Wang-Wu	0.58728	17.458	1	0.0%	—
Dornaika-Horaud	0.585795	10.0628	1	0.0%	—
Zhuang-Roth-Sudhakar	0.590747	10.132	1	0.0%	—
Calibrex nonlinear refinement	0.58721	10.0394	1	0.0%	—

Calibrex's nonlinear refinement is best on translation in this benchmark; Shah is marginally best on rotation. These are held-out closure errors, not ground-truth extrinsic errors. Read the benchmark protocol and provenance.

Integration paths

• Open3D SLAC adapter
• Targetless LiDAR-camera adapter
• Problem builder
• Schema reference
• License boundaries

Current status

Calibrex is alpha research software. LiDAR-LiDAR and hand-eye paths have native solvers, evaluation, and public examples. Camera-LiDAR, radar, RGB-D, and some online paths remain experimental. Failed controls are reported rather than hidden; see the changelog and development roadmap for current limitations.

Regenerate committed schemas with calibrex schema all --output-dir schemas.